Computer Science Review 57 (2025) 100747

Contents lists available at ScienceDirect

Computer Science Review

journal homepage: www.elsevier.com/locate/cosrev

Review article

Review of Distributed Quantum Computing: From single QPU to High

Performance Quantum Computing
David Barral a , F. Javier Cardama b , Guillermo Díaz-Camacho a , Daniel Faílde a , Iago F. Llovo a ,
Mariamo Mussa-Juane a , Jorge Vázquez-Pérez a,b , Juan Villasuso a , César Piñeiro c ,
Natalia Costas a , Juan C. Pichel b,c , Tomás F. Pena b,c ,∗, Andrés Gómez a
a
Galicia Supercomputing Center (CESGA), Avda. de Vigo S/N, Santiago de Compostela, 15705, Spain
b
Centro Singular de Investigación en Tecnoloxías Intelixentes (CiTIUS), Universidade de Santiago de Compostela, Santiago de Compostela, 15782, Spain
c
Departamento de Electrónica e Computación, Universidade de Santiago de Compostela, Santiago de Compostela, 15782, Spain

ARTICLE INFO ABSTRACT

Keywords: The emerging field of quantum computing has shown it might change how we process information by using
Distributed quantum computing the unique principles of quantum mechanics. As researchers continue to push the boundaries of quantum
High-performance computing technologies to unprecedented levels, distributed quantum computing raises as an obvious path to explore with
Teleportation
the aim of boosting the computational power of current quantum systems. This paper presents a comprehensive
Quantum networks
survey of the current state of the art in the distributed quantum computing field, exploring its foundational
Distributed quantum compilers
Circuit knitting
principles, landscape of achievements, challenges, and promising directions for further research. From quantum
Distributed quantum applications communication protocols to entanglement-based distributed algorithms, each aspect contributes to the mosaic
of distributed quantum computing, making it an attractive approach to address the limitations of classical
computing. Our objective is to offer a comprehensive review that serves both experts in the field and
researchers or enthusiasts in quantum computing looking for a starting point to explore the area of distributed
quantum computing.

Contents

1. Introduction ...................................................................................................................................................................................................... 2
2. Physical layer for distributed quantum computing ................................................................................................................................................ 3
2.1. Quantum entanglement........................................................................................................................................................................... 3
2.2. Quantum teleportation or teledata ........................................................................................................................................................... 3
2.3. Variants of quantum teleportation ........................................................................................................................................................... 4
2.3.1. Entanglement swapping ............................................................................................................................................................ 4
2.3.2. Quantum gate teleportation or telegate ...................................................................................................................................... 5
2.3.3. Multipartite teleportation .......................................................................................................................................................... 5
2.4. Quantum devices for entanglement distribution ........................................................................................................................................ 5
2.4.1. Quantum transducers................................................................................................................................................................ 5
2.4.2. Quantum memories .................................................................................................................................................................. 6
2.4.3. Quantum repeaters ................................................................................................................................................................... 7
2.4.4. Entanglement routers and switches ............................................................................................................................................ 7
3. Networks for distributed quantum computing....................................................................................................................................................... 7
4. Development layer ............................................................................................................................................................................................. 9
4.1. Types of distribution............................................................................................................................................................................... 9
4.1.1. Circuit distribution ................................................................................................................................................................... 9
4.1.2. Circuit cutting.......................................................................................................................................................................... 12
4.1.3. Embarrassingly parallel............................................................................................................................................................. 13
4.2. Compilation ........................................................................................................................................................................................... 14

∗ Correspondence to: Centro Singular de Investigación en Tecnoloxías Intelixentes (CiTIUS), Universidade de Santiago de Compostela, 15782 Santiago de
Compostela, Spain.
E-mail address: [email protected] (T.F. Pena).

https://doi.org/10.1016/j.cosrev.2025.100747
Received 10 April 2024; Received in revised form 24 February 2025; Accepted 9 March 2025
Available online 24 March 2025
1574-0137/© 2025 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
D. Barral et al. Computer Science Review 57 (2025) 100747

4.2.1. Analysis phase ......................................................................................................................................................................... 14

4.2.2. Distributed quantum intermediate representation ........................................................................................................................ 15
4.2.3. Synthesis phase ........................................................................................................................................................................ 15
4.2.4. Available compilers .................................................................................................................................................................. 17
5. Application layer ............................................................................................................................................................................................... 19
5.1. Circuit-distribution based applications ...................................................................................................................................................... 20
5.2. Circuit cutting and other hybrid applications ............................................................................................................................................ 20
5.3. Embarrassingly parallel applications......................................................................................................................................................... 21
6. Final remarks and open challenges ...................................................................................................................................................................... 21
Declaration of Generative AI and AI-assisted technologies in the writing process .................................................................................................... 22
Declaration of competing interest ........................................................................................................................................................................ 22
Acknowledgments .............................................................................................................................................................................................. 22
Data availability ................................................................................................................................................................................................ 22
References......................................................................................................................................................................................................... 22

1. Introduction

In pursuing superior computational abilities, quantum computing

has emerged as a promising frontier with huge potential. While individ-
ual quantum systems have shown impressive capabilities, distributed
quantum computing introduces a new approach that could vastly in-
crease computational power. This study aims to explore in depth the
current landscape of DQC, also known in certain literature as modu-
lar quantum computation, from physical devices and interconnection
networks to distributed algorithms. In this review, we will analyze the
different solutions proposed and the challenges posed by this rapidly
advancing field.
As we examine distributed quantum systems more closely, it be-
comes clear that collaborative and interconnected quantum processors
are essential for overcoming the constraints faced by standalone sys-
tems. Problems of both fundamental origin – decoherence, dissipation,
and crosstalk – and practical origin – processor topology, cabling,
connectors, and control electronics – hinder the fabrication of ultra-
large Quantum Processing Units (QPUs) [1]. It is thus foreseeable in the
short term that quantum computers will not scale in a local device with
a large number of qubits in a single quantum processor. A distributed
infrastructure with several quantum processors that contain a limited
number of qubits could overcome this difficulty. In fact, there is a clear Fig. 1. Layered model for distributed quantum computing.
consensus among leading academic and industry stakeholders that the
practical realization of large-scale quantum processors should adopt a his famous criteria for a quantum computer, two additional not-so-
distributed approach based on clusters of small, modular quantum chips well-known items related to DQC and the interconnection of QPUs:
within a network infrastructure, with classical and/or quantum commu- the ability to interconnect stationary and flying qubits and to transmit
nications [2–4]. QPUs are intended to be seamlessly integrated into a flying qubits between specified locations faithfully.
classical High-Perfomance Computing (HPC) infrastructure, alongside After the first theoretical studies on the feasibility of DQC, a series of
CPUs, GPUs, and other hardware accelerators [5–9]. This integration proposals for experimental realizations began to appear gradually [18–
allows for their utilization in collaboration within a shared develop- 21]. At the same time, several interesting developments regarding DQC
ment environment, leading to what is already called quantum-centric algorithms were made, such as the distributed versions of the Grover
supercomputing centers [10]. and Shor algorithms [22,23]. The first taxonomy of DQC systems was
As an example of this trend, IBM recently unveiled Quantum System proposed by Yepez [24] in the early 2000s, where two types of systems
Two [11], a modular architecture that will serve as the basis for were described: those with entanglement between nodes, called type-I,
building their new quantum-centric HPC infrastructures. The model and those with only inter-node classical communication, called type-II.
unveiled features three IBM Quantum Heron processors, each with 133 Jozsa and Linden later demonstrated that a type-II quantum computer
fixed-frequency qubits and tunable couplers. According to IBM, Heron cannot achieve exponential speedup when the computation requires
yields a 3-5x improvement in performance with respect to the previous entanglement across the full set of qubits [25].
127-qubit Eagle processor, virtually eliminating crosstalk. Considering these initial works as a starting point, this review
However, the interest in DQC is not new. We have to go back extensively examines the current advancements in the field of DQC,
to the end of the 20th century to find the first works that analyzed extending and updating previous surveys on this subject. The previ-
the possibility of using non-local effects to perform distributed com- ous1 survey by Caleffi et al. [26] provides a comprehensive overview
puting [12,13]. This interest grew after Cirac et al.’s work, where it of DQC archetypes. It covers a range of configurations, from sin-
was shown that DQC is superior to classical computing for the phase gle QPU with multi-core execution to single datacenter with multiple
estimation problem even under non-ideal conditions [14]. Shortly after,
Eisert et al. [15] and Collins et al. [16] took a step forward, introducing
resource-optimized protocols for non-local quantum gates necessary 1
During the review process of this article, we were notified that an updated
to move from specific problems like phase estimation to universal version of the Caleffi et al. survey available in arXiv was published in the
quantum computing. At the same time, DiVincenzo [17] included, in Computer Networks journal.

2
D. Barral et al. Computer Science Review 57 (2025) 100747

QPUs and even multi-farm networks, also comparing them to their teleportation or telegate. Furthermore, the basic two-node teleportation
classical counterparts. Additionally, it offers a detailed section on simu- can be extended to multi-party distribution networks composed of
lation tools for DQC development, organized into three distinct layers: many nodes. Some parties may either help the rest of the network in
hardware-oriented, protocol-oriented, and application-oriented. the quantum communication protocol – assisted teleportation –, or the
Our review is specifically focused on the integration of HPC with quantum information may be imperfectly broadcast from one sender to
DQC, and we excluded multi-farm networks due to the substantial la- the rest – quantum telecloning.
tency introduced by geographically separated centers. Instead, we con- In the following sections, we will introduce these protocols in detail.
centrate on multi-core and multi-QPU configurations, drawing parallels
with classical HPC paradigms. For example, we explore techniques such 2.1. Quantum entanglement
as circuit cutting and embarrassingly parallel distribution of circuits,
whenever fully realized quantum networks may not yet be available. Entanglement is the property of a quantum system that illustrates
Additionally, we emphasize the critical importance of the physical layer the impossibility of describing a composed system in terms of just
in DQC systems. its individual components due to nonclassical correlations of certain
To facilitate the readers’ understanding, this survey is structured degree(s) of freedom of the subsystems [30]. Typical examples of these
according to a layered model, as depicted in Fig. 1, similar to the full- degrees of freedom are the position and momentum of free particles,
stack architecture presented by [27], the abstract model in [28], or the the polarization of light, energy levels of trapped ions, or transverse
DQC simulator tools structure in [26]. atomic spins. These degrees of freedom are related to observables that
The two lower layers in Fig. 1 encompass the hardware develop- present a discrete and finite spectrum or a continuous and infinite one.
ments needed to implement a distributed quantum system and would Hence, the terms discrete variable (DV) and continuos variable (CV).
be equivalent to the three lower layers of the classical OSI model. So, This review focuses on DV because it is the most common in quantum
the physical layer refers to the mechanisms that allow two physically computing.
separated QPUs to be connected, while the network layer defines how Archetypical examples of DV entangled quantum states are the pure
to establish communication between multiple QPUs. Directly above this states
layer, we discuss advances in development tools that allow applications 1 ( )
|𝛷± ⟩ = √ |0⟩𝐴 |0⟩𝐵 ± |1⟩𝐴 |1⟩𝐵 ,
to be distributed and executed on a distributed quantum system, includ-
2
ing partitioning, compilation, optimization, and mapping algorithms. (1)
1 ( )
Finally, in the uppermost layer, we address distributed algorithms. It |𝛹 ± ⟩ = √ |0⟩𝐴 |1⟩𝐵 ± |1⟩𝐴 |0⟩𝐵 ,
is important to note that these layers are interdependent, with each 2
layer influencing those immediately preceding and succeeding. For dubbed Bell states or Einstein, Podolski and Rosen (EPR) pairs, where
instance, the development of a compiler is influenced by the underlying two parties – Alice and Bob – share two qubits A and B encoded in a
hardware and provides support for different partitioning techniques in dichotomic degree of freedom as polarization, spin, or any other two-
the application layer. level quantum variable [31]. A perfect non-local correlation arises as
Following this structure, the review is organized as follows. Sec- Alice’s measurement outcome determines Bob’s measurement outcome.
tion 2 describes the available quantum mechanical tools to transmit This property allows us to build an intuition of how Bell states are a
quantum information. In Section 3, we present proposals oriented natural choice for quantum communication: if a quantum gate, whose
to creating networks interconnecting multiple QPUs. Next, Section 4 matrix representation is symmetric, is applied to one of the qubits of
discusses solutions that allow applications to run in distributed environ- the Bell state |𝛷+ ⟩, it is the same as if the gate was applied to the
ments, including partitioning, distribution, compilation, and mapping other qubit. The gate somewhat ‘slides’ between qubits through the
techniques. Section 5 presents different proposals for applications run- entanglement, like beads on a string [32].
ning in these environments. We will end the paper with a summary of These entangled states are the basis of a large number of quantum
the current state of the art and open lines in the field. information protocols, one of which is quantum teleportation, which
we introduce in the following section.
2. Physical layer for distributed quantum computing
2.2. Quantum teleportation or teledata
DQC aims at performing arbitrary computational tasks between
unknown quantum states at the distant nodes of a quantum network. Quantum teleportation, one of the more remarkable quantum in-
These networks, identically to their classical counterparts, coordinate formation protocols, was introduced thirty years ago in a landmark
and distribute information across devices. However, quantum networks paper [33]. This quantum protocol enables the reconstruction of an
have multiple features and limitations that make these tasks difficult, unknown quantum state of a given physical system at a different loca-
primarily arising from the no-cloning theorem: arbitrary quantum states tion without actually transmitting the system. Quantum teleportation
cannot be perfectly copied; therefore, quantum information cannot be requires two key ingredients: Quantum entanglement and classical
replicated and broadcast [29]. Fortunately, the properties of quan- communication between the locations (which excludes superluminal
tum systems can be exploited in a way that allows us to circumvent communication).
this impediment and reliably transmit quantum information or control Quantum teleportation plays a pivotal role in the development
quantum systems remotely. This section will briefly describe which of quantum technologies [34]. It overcomes some of the limitations
quantum mechanical tools are available for this purpose. of quantum communications and quantum computing using the non-
First and foremost, the physical resource that enables performing local transfer of unknown information. Quantum teleportation net-
non-local computation is entanglement, a unique correlation of joint works [35], entanglement swapping [36], and quantum repeaters [37]
quantum systems stronger than any classical counterpart but very frag- enable the distribution of entanglement over long distances [38], while
ile, hard to create and to maintain long. Entanglement lies at the heart quantum gate teleportation [39] and measurement-based quantum
of quantum communications, facilitating the distribution of quantum computing [40] are examples of techniques that distribute local gate
states encoding quantum information through a protocol known as operations among physically disconnected parties [41]. Entanglement
quantum teleportation or teledata. Multiple teleportation variants exist, swapping and gate teleportation will be discussed further in the next
which are designed to either transmit data in one direction – quan- section.
tum teleportation or teledata – but also bi-directional communication Proof-of-principle demonstrations of quantum teleportation were
– entanglement swapping – and gate operation at a distance – gate successfully achieved using diverse physical substrates as photonic

3
D. Barral et al. Computer Science Review 57 (2025) 100747

Table 1
Some milestones in quantum teleportation in terms of Bell efficiency, fidelity, distance
of teleportation, and quantum memory. QED: quantum electrodynamics.
Quantum technol. Bell eff. Fidel. Max. dist. Memory
Polarization [38] 25% 0.80 1400 km NA
Integrated opt. [51] 25% 0.894 10 m NA
Superconduct. [41] 100% 0.79 chip 1 ms
Cavity QED [64,65] 100% 0.833 60 m –
Ion Trap [66] 100% 0.845 chip –
Rare-earth [67] 50% 0.86 1 km 17.5 μs

varies for different information encodings: for instance, for a simple

realization of Bell-state measurement using DV photonic qubits, the
Bell efficiency is 50% at maximum [55].
2. The teleportation fidelity 𝐹 ∈ [0, 1] between the input state 𝜌 and
Bob’s output state averaged over all Alice’s measurement results
and input states. The benchmark for the teleportation fidelity is
surpassing the fidelity for state transfer without quantum resources,
using, for instance, just classical correlations, i.e., 𝐹 > 𝐹class , where
𝐹class = 2∕3 for DV [56].

Table 1 shows examples of recent milestones in quantum teleporta-

tion in different technologies. More details on the state of the art can
be found in [57,58].
Quantum teleportation has seamlessly made the leap from labora-
tory conditions to real-world implementation in urban environments,
showcasing its adaptability and robust functionality. Teleportation net-
works allow for the reliable transfer of quantum information between
a number of distant nodes, even in the presence of non-ideal features
such as noise and loss. Recent advances include demonstrations of two-
node teleportation over a metropolitan network [59,60], links between
nanophotonic memories and ion traps in an urban network [61,62], and
multinode entanglement over a metropolitan network with a cloud of
Rubidium atoms in a ring cavity acting as a quantum memory [63].
More on quantum networks will be delved in Section 3.

Fig. 2. Sketch of quantum communication protocols: (a) Quantum-state teleportation

2.3. Variants of quantum teleportation
(teledata), (b) entanglement swapping, and (c) quantum-gate teleportation (telegate).
BSM: Bell-state measurement. CM: controlled operation and projective measurement.
Quantum teleportation is a primitive of quantum information sci-
ence and has a number of variants essential for DQC. In the following,
we review the most important three: entanglement swapping, quantum
qubits [42], optical modes [43], atomic ensembles [44], nuclear mag-
gate teleportation – telegate – and multipartite teleportation.
netic resonance [45], trapped atoms [46,47], and solid-state systems
[48]. Over the last years, the focus has moved to teleporting more
2.3.1. Entanglement swapping
complex states – larger number of degrees of freedoms or higher
Entanglement swapping is a variant of quantum teleportation that
dimension qubits [49,50] – and to real-world applications in quantum
enables remote correlations by the transfer of quantum entanglement
communications and computation [38,51,52].
between distant end-users that do not directly share a quantum re-
In the teledata protocol, Alice and Bob share an entangled Bell state
source. In this case, Bob shares two entangled states, one with Alice
as that given by Eq. (1) [42], see Figs. 2(a) and 3(a) in physical and and the other with Charlie, as shown in Fig. 2(b). Bob acts as a relay
circuit representations, respectively. A third party provides Alice with a between them, performing Bell measurements and broadcasting the
qubit C to be teleported to Bob. Importantly, the quantum state of qubit outcomes by a classical channel to them, who apply the suitable gates to
C –represented by 𝜌– is unknown to both Alice and Bob unlike in remote their qubits. As a result, Alice and Charlie now share an entangled state
state preparation [54]. Alice then performs a Bell-state measurement conditioned on the result of Bob’s measurement [36]. This protocol,
(BSM), which randomly projects with equal probability her qubits A together with entanglement distillation2 [68], enables the distribution
and C into one of the four Bell states |𝛷± ⟩ or |𝛹 ± ⟩. As a result, Bob’s of entanglement over large distances, being the basis of quantum
qubit B is simultaneously projected onto the state 𝑇 † 𝜌𝑇 , where 𝑇 ∈ repeaters [37]. Related to entanglement swapping are fusion gates [69,
{𝐼, 𝑋, 𝑍, 𝑍𝑋} is an elementary or a combination of Pauli operators. 70], where projective measurements probabilistically fuse small entan-
As the last step, Alice informs Bob of the BSM outcome through the gled states in order to produce large entangled states – cluster states –
classical channel using two classical bits – feed-forward – and Bob useful for measurement-based quantum computing [40].
applies the suitable gate 𝑇 to his qubit to recover the unknown state 𝜌 The first demonstration of entanglement swapping was carried out
of qubit C at his location. by Pan et al. using polarization-entangled photons [71]. Swapping
Regarding the figures of merit of quantum teleportation, there are
mainly two:
2
Entanglement distillation, aka entanglement purification, involves con-
1. The BSM efficiency or Alice’s success probability for distinguishing verting 𝑁 copies of any entangled state 𝜌 into a certain quantity of nearly
a complete basis of entangled states – like the four Bell states. This pure Bell pairs, solely through local operations and classical communication.

4
D. Barral et al. Computer Science Review 57 (2025) 100747

Fig. 3. Examples of teledata and telegate circuits for the application of CZs gates over |𝑡1 ⟩ and |𝑡2 ⟩ with the remote state |𝑎⟩ as control. (a) The state |𝑎⟩ in QPU1 is teleported to
the first qubit of QPU2 (b) Cat-entangler and cat-disentangler primitives [53] are used to implement the remote control.

has been recently applied to connect two spatially-separated solid- 2.4. Quantum devices for entanglement distribution
state quantum memories by telecom links [67], and to entangle non-
neighboring Nitrogen Vacancy (NV) qubits in a multinode teleportation In the search for maximum performance and demonstrating quan-
network [72]. tum advantage for distributed, scalable quantum computing systems,
modular architectures featuring specialized, single-purpose hardware
2.3.2. Quantum gate teleportation or telegate are currently under development [81]. The quantum devices that are
In gate-based quantum computing, a sequence of unitary operations part of these architectures, apart from QPUs, can be categorized in
(usually single- and two-qubit) are applied on a set of qubits. However, one of the following categories: quantum transducers, quantum memories,
sometimes there is no direct interaction between qubits on which we quantum repeaters, and entanglement routers and switches. This section
want to apply a two-qubit gate [20]. Quantum gate teleportation, also
will describe the aforementioned devices in detail and discuss the
known as telegate, reduces the topological requirements by substituting
current research advances in each technology.
two-qubit gates with other cost-effective resources: auxiliary entangled
This section will detail the aforementioned devices in detail and
states, local measurements, and single-qubit operations [39]. Typically,
discuss the current research advances in each technology.
Alice and Bob want to perform a non-local operation on unknown
control and target states using a shared Bell state as a quantum channel.
To this end, both perform locally controlled operations and projective 2.4.1. Quantum transducers
measurements (CM) on their half Bell state and control/target states. The communication between local qubits of systems where the
After this step, partial quantum information is transferred between the quantum operations take place (e.g., QPUs, memories or repeaters)
two parties conditioned to the measurement outcomes. Cross communi- requires the conversion, or transduction, of their states to a different
cation of the results through a classical channel enables Alice and Bob system used for delivery of quantum states in the form of flying qubits,
to perform suitable corrections to the control and target states. This which have the requirements of being highly mobile and well coupled
procedure results in a controlled gate operation on two non-interacting to the specific local platform. Multiple flying qubit systems have been
input states – see Figs. 2(c) and 3(b) for physical and circuit represen- proposed, such as short-distance electronic states in semiconductor
tations, respectively. The first experimental demonstration of quantum devices [82], direct delivery of nuclei with long-lived nuclear-spin
gate teleportation was a remote CNOT operation carried out through qubit encoding [83] and, more commonly, single photons, given their
photon entanglement and linear optical manipulations [73]. Recent ad- naturally mobile nature and their low coupling with the environment.
vances in remote operations comprise superconducting qubits, trapped In classical communications, the high-rate transfer of current tech-
ions, and quantum electrodynamics cavity nodes [41,64,66]. nologies is only possible due to the high bandwidth, and low atten-
When applied to multipartite entangled states with a given topol- uation and latency provided by light in fiber optics, enabling the
ogy, suitable measurement on a given network node teleport unitary- underwater connection of continents at tens of thousands of kilo-
transformed-state to other nodes. This is the basis of measurement- meters [84]. The current state-of-the-art telecommunication systems
based quantum computing [40]. also implement multiplexing, i.e., encoding information at multiple
wavelengths through the same fiber [85]. As DQC requires determin-
2.3.3. Multipartite teleportation istically distributing entanglement, the requirements for flying qubits
Multipartite entangled states as the Greenberger-Horne-Zeilinger
are, primarily, good coupling to the particular local quantum system,
(GHZ) state enable a natural extension of quantum teleportation to
either by direct emission or interaction, and that the fidelity of the
more than two parties [74]. These 𝑁-party protocols for multipartite
resulting entangled states is maximal. Quantized states of light are
teleportation enable two variants: assisted and unassisted teleportation
also the most natural information carrier choice for the distribution of
– commonly referred to as quantum telecloning. In the first case,
quantum states at a distance, and extensive research has focused on
assisted teleportation, Alice helps the communication between Bob and
Charlie by performing a tailored measurement and broadcasting the the accurate manipulation of photonic states using linear and nonlinear
result to them, thus improving the entanglement between them [35]. optical devices [86,87].
In the second case, quantum telecloning, Charlie teleports to Alice and The most widely studied way of generating entanglement between
Bob simultaneously, hence with a teleportation fidelity, limited by the remote systems is via entanglement swapping between two indepen-
no-cloning theorem, given by 𝐹 = (𝑀𝑁 + 𝑀 + 𝑁)∕(𝑀𝑁 + 2𝑀), for 𝑁 dently entangled flying qubit-matter qubit systems, i.e., generating
senders and 𝑀 receivers of qubits [75]. entanglement between photons and local qubits (trapped ions, neu-
Examples of assisted teleportation are open-destination teleporta- tral atoms, or NV centers), then performing BSM on the photons of
tion [76] and, more recently, shared-quantum-secret teleportation [77]. each pair. Hence, their joint wavefunction collapses in the same non-
Quantum telecloning was, in turn, demonstrated in DV by means of separable state, and the matter systems become entangled. To this
partial teleportation [78]. Cloning of entanglement [79] and copy purpose, heralded entanglement of photons emitted after de-excitation
distribution [80] are recent examples of this variant of teleportation. from prepared excited states has been shown in trapped-ion qubits [88–

5
D. Barral et al. Computer Science Review 57 (2025) 100747

90], neutral atoms [91] and diamond NV-center qubits [92–96]. After not lead to a mixed state in the matter qubits, so it can be tried again.
the subsequent BSM, fidelities to Bell states of up to 88% at 230 m While it is a new and promising avenue for the distribution of quantum
have been demonstrated in trapped-ions [97], and even above 60% states, further research is required to bring its quantum efficiency and
between NV-centers separated by a 25 km metropolitan fiber link [98]. fidelity closer to unity.
Deterministic qubit state transfer between different NV-center nodes Moreover, interconnecting quantum systems may require coupling
has also been achieved [72]. One promising proposal is the coupling platforms that operate at different photon frequencies. For this purpose,
of ion- or Rydberg atom-chains in optical cavities [99], which has techniques are being developed to implement frequency conversion
been shown capable of providing any-to-any entanglement for large of single photons on demand, maintaining certain properties (such
systems with over 500 qubits in trapped ions by using two atomic as polarization) intact, which would enable the transcoding of qubits
species, one of which acting as a communication qubit and another as between platforms. One such technique is heralded up-conversion from
memory qubit. Quantum dots are also promising due to their tunable infrared to visible light, which has been achieved through sum fre-
emission wavelengths in the infrared range, yet some challenges remain quency generation in nonlinear crystals [114,115]. More recently, Mu-
such as extending the qubit lifetime [100,101]. Rare-earth doped crys- rakami et al. [116] have demonstrated frequency conversion from
tals, commonly Eu3+ - or Pr3+ -doped Y2 O3 crystals also have emission visible to infrared using pairs of non-degenerate photons generated
near the 1550 nm band with sharp linewidths and long coherence by SPDC, and Weaver et al. [117] have shown frequency bidirec-
times [102]. Spin-photon coupling in the microwave range has also tional transduction from microwave to infrared light using transduction
been demonstrated in Si double quantum dot spin qubits by coupling assisted by a resonant mechanical mode. However, the quantum effi-
the charge dipole of a trapped electron to the electric field component ciency of these techniques is currently low and significant efforts are
of a cavity photon stored in a superconducting resonator [103]. The underway to push it towards unity. In addition to the aforementioned
promising advantages of these links is that multiple quantum dot qubits frequency conversion techniques, recent work by Sahu et al. [118]
can use the same resonator, as their coupling can be switched on and has demonstrated deterministic entanglement between the quadratures
off in nanoseconds, and the resonators’ dimensions are large compared of propagating microwave and optical photons in cryogenic waveg-
to the double quantum dots. uides, a first step towards interconnecting superconducting qubits with
Finding mechanisms to link superconducting chips together to over- long-range communication systems and memories.
come their scaling needs is a current technological challenge. One of the In summary, there are multiple competing techniques which allow
most interesting experiences has been the deterministic transmission of distant QPU to generate entanglement in virtually all matter qubit
excitations between superconducting QPUs using cryogenic microwave technologies. However, much research is still required to push both
waveguides [104,105]. Applying modulated microwave pulses, an ef- fidelity and efficiency towards unity. Frequency conversion is a promis-
fective coupling between energy levels in the transmon qubits and ing technique which may allow future interfacing of different systems,
their respective transfer resonators can be achieved, transferring an enabling heterogeneous DQC.
excitation from node A to its transfer resonator, which then emits a
single microwave photon towards node B where it is absorbed, exciting 2.4.2. Quantum memories
its qubit. Cryogenic, lattice-based quantum networks have been pro- To fully take advantage of the entanglement distribution and distil-
posed based on this method of connecting superconducting chips [106], lation protocols for both short and long distance quantum communi-
demonstrating high fidelity, in excess of 85% when correcting for cation, it is paramount that the coherence time of the communication
readout errors. However, some serious drawbacks remain, such as the qubits is longer than the protocol itself, surviving multiple rounds of
high cost of cryogenic equipment, complexity and low modularity of qubit exchange and entanglement purification. These long-lived qubits,
these systems. organized as large registries, are known as quantum memories or
Correlated photon sources such as spontaneous parametric down- quantum Random Access Memorys (qRAMs).
conversion (SPDC) or quantum dots can also be utilized to achieve the The simplest quantum memories are photonic memories, in which
initial photon–matter qubit entanglement. SPDC sources consist of a photons are stored and then retrieved after a given time. Multiple
non-linear crystal pumped by a strong laser beam generating pairs of approaches exist, such as using free space optical loops triggered by
maximally entangled photons with some probability, which can then heralding [119] or fiber delay lines [120] and cavities with tunable
be frequency-filtered and used to interact with the physical qubits. Q-factor [121,122]. Stimulated photon-echo is a more advanced tech-
Using this technique, distant solid-state quantum memories have been nique based on the absorption and delayed reemission of single pho-
entangled at distances of 1 km and fidelities of up to 86% using two tons with the same quantum state after an ensemble of atoms is
photons at 606 nm and 1550 nm [107]; the former is stored in the rephased [123–125], which has been demonstrated e.g., using slow
collective excitation of Pr3+ in a doped crystal using the Atomic Fre- light by electromagnetically-induced transparency (EIT) [126], con-
quency Comb (AFC) protocol, while the latter is sent to a BSM analyzer trolled reversible inhomogeneous broadening (CRIB) [127] and atomic
and corrected, resulting in entanglement. Hyperentanglement, where frequency combs (AFC) in rare-earth doped crystals [102,128,129].
more than one degree of freedom can simultaneously be maximally All-photonic systems (i.e., photonic quantum computing) can already
entangled (e.g., polarization and direction of two photons) has also take advantage of photonic memories, as they do not require transduc-
been demonstrated using this type of sources [87,108]. Alternative tion [130,131].
quantum dot-based sources have very attractive properties for this However, both the difficulty of retrieving single photons with high
purpose, such as being triggered on-demand and energy-tunable [109– fidelity as well as the low scalability of photonic-based memories have
111], and reaching fidelities over 90% [112,113]. pushed forward extensive research on multiple alternative quantum
A less widely studied possibility is making single photons interact memory technologies, demonstrating high-fidelity single-qubit gates in
in-flight with two separated quantum systems. Using ancillary doubly excess of the threshold needed for quantum error correction [132,133].
reflected single photons followed by a measurement of the photon and Notable examples are trapped-ion and -neutral atom qubits, which use
a conditional rotation of the target qubit, heralded deterministic tele- the hyperfine structure of atomic ensembles of ions [134], or neutral
data [65] and telegate entanglement of remote qubits [64] have been alkali or alkaline earth single atoms in optical tweezers [135–137] to
demonstrated. Achieving fidelities up to 90% and 85% respectively at encode the quantum states, which can be individually addressed by
60 m, this technique could enable deterministic, short-distance, low- microwave pulses [138]. Quantum memories based on diamond NV-
latency DQC. Its advantages are two-fold: teleportation is performed centers have also been demonstrated (see [139] and references therein).
without the need for preformed Bell pairs, so it can be done just-in- Some of these technologies have demonstrated long coherence times,
time; and losing the photon (i.e., not being able to measure it) does of up to 10 min in single trapped-ion qubits [140] and up to six hours

6
D. Barral et al. Computer Science Review 57 (2025) 100747

in cryogenically cooled Eu3+ -doped yttrium orthosilicate nuclear spin (IP), while switches only recognize which physical addresses are routed
qubits [83]. More recently, Barnes et al. [137] have demonstrated through their connections to redirect traffic. The current absence of a
an individually addressable 21-qubit register of highly coherent and quantum IP standard makes the distinction of the quantum counterparts
independent qubits with coherence times of about 40 s using nuclear difficult, so authors have been using these terms interchangeably. More-
spin qubits in optical tweezers, opening the gate to intermediate-scale over, the quantum hardware required is essentially the same and any
quantum memories. differences would arise from the higher-level classical network manage-
ment. Following this description, any two QPUs in the network can be
2.4.3. Quantum repeaters connected through either one or multiple switches and/or routers in a
As we previously discussed, light is the most natural long-distance Quantum Local Area Network (QLAN), or through an efficient routing
carrier of quantum states. However, the absorption of light imposes path that connects multiple routers (which may require repeaters to
intrinsic physical limits on the distance at which single photons can maintain entanglement) and lead to a Quantum Wide Area Network
travel. In long-distance fiber communications, absorption is mainly (QWAN) [106,156]. The interconnection of quantum networks could
produced by the fiber, with an attenuation coefficient in the range eventually lead to a worldwide Quantum Internet [157,158]. However,
∼0.14−0.4 dB/km in low loss telecom fibers [141,142]. Furthermore, this escapes the scope of this review [156,159,160].
even in the short-distance communication range of a datacenter, the Entanglement switches and routers can then be thought of as single-
rate at which photons are lost is nontrivial: the typical loss per SC con- purpose QPUs: their sole objective is establishing entanglement among
nector is ∼0.25 dB [143], so the shortest possible connection between compute nodes through entanglement swapping, for which implement
two nodes accounts for ∼0.5 dB of attenuation, i.e., ∼11% of the photons all the quantum technology required, such as quantum registries, entan-
are lost. Hence, if frequent quantum communication is required for a glement sources and means to perform BSM, as well as all the hardware
distributed algorithm, the error probability quickly increases as 𝑒 = required for networking logic and classical communications [160].
1−10𝑛⋅dB∕10 after 𝑛 exchanges, limiting the scalability and reliability of Moreover, these devices may also be built on different quantum plat-
the calculation. forms than the proper QPUs, e.g., not requiring the implementation
It is important to understand that any improvements in the connec- of a complete set of quantum gates but only those required for the
tor losses and fiber attenuation cannot and will not solve the problem swapping protocol and instead requiring registries of qubits with very
of exponential decay with 𝑛. Given that standard telecommunications high fidelity and coherence times longer than the entanglement distil-
erbium-doped fiber amplifiers (EDFA) cannot be used to amplify ar- lation protocol, or access to quantum memories that fulfill these two
bitrary quantum states due to the no-cloning theorem, quantum re- requirements. Some proposals suggest networks based on single atoms
peaters are essential to the implementation of entanglement distribution trapped and coupled to optical resonators as memory qubits, which
have long coherence times and good photon coupling (see [161] and
and teleportation which enable deterministic transmission of quantum
references therein).
states and remote quantum operations between nodes [144,145]. An
early solution to the problem of implementing a quantum repeater was
3. Networks for distributed quantum computing
proposed by Briegel et al. [37], which consisted of first entangling noisy
and imperfect qubits and then creating a high-fidelity entangled pair
The scientific literature on quantum networks is really extensive and
through entanglement distillation. Recent proposals have extended the
a dedicated review would be needed to properly address all aspects re-
idea of entanglement distillation to qudits (i.e., 𝑑-state systems) [146],
garding architecture, entanglement creation and distribution, network
multiple simultaneously entangled degrees of freedom (hyperentangle-
orchestration, network software stack and protocols. Therefore, our
ment) [147,148], and logical qubits [114,149]. Van Leent et al. [150]
intention in this section is to provide sufficient information to give the
have demonstrated single-atom entanglement over a 33 km telecom
reader an outlook of some relevant progress on this subject. Although
fiber using quantum repeaters, proving that long distance entanglement considerable amounts of research has been oriented towards communi-
is already a technical possibility. Recent work has also shown that cation systems for the Quantum Internet, much of it can be applied
Er3+ inclusions in calcium tungstate greatly diminish optical spectral to DQC. However, in DQC, the focus should be on short/datacenter
diffusion [151], a requirement to generate indistinguishable single distance limits.
photons needed for optical repeaters, as this ion is well coupled by its Quantum networks (QNs) enable the execution of distributed op-
telecom band optical transition. erations among two or more qubits that may be very close to each
other or separated by long distances. The mechanisms used for com-
2.4.4. Entanglement routers and switches munication could be based on the transmission of the quantum states,
As previously explained, the execution of general quantum algo- or on the creation, distribution and consumption of entanglement. The
rithms in multiple qubit-limited QPUs requires entanglement to be entanglement resources provided by these QNs can be used both in
generated on demand between pairs of arbitrary qubits [152]. For DQC and in other applications of quantum technologies e.g., sensing
this reason, recent research has focused on implementing teleportation or encryption. A comprehensive review on entanglement networks
protocols between non-neighboring nodes. The simplest way to obtain covering the fundamental mechanics, the enabling technologies, the
arbitrary entanglement with interconnected QPUs is pre-establishing network architecture and elements and research challenges associated
shared entanglement, as discussed in Section 2.4.1, in a one-to-one can be found in [162].
fashion between specific communication qubits in different nodes. In Entanglement networks allow the execution of the previously dis-
these one-to-one schemes, not every pair of QPUs ought to be physi- cussed swap, teleport or telegate mechanisms. Besides these well known
cally connected, reducing the complexity of implementation for small network mechanisms, Miguel-Ramiro et al. propose the inclusion of
integrated systems. full quantum functionalities that increase the parallelism of opera-
However, this apparent simplicity faces a major scalability chal- tions using superposition of tasks with quantum control [163]. Exam-
lenge, resulting in substantial qubit swap and distillation overhead ples of this are the execution of superposed tasks such as superposed
in complex, strongly entangled algorithms [14]. While compilation measurement/non-measurement, superposed paths, superposed tele-
optimizations can reduce swap operations, more general and mod- porting/non-teleporting or superposed merging/non-merging of graph
ular quantum networks will need entanglement routers and switches states.
to distribute entanglement between arbitrary qubits, akin to classical Classical network architectures and protocols cannot be directly
counterparts [153–155]. extrapolated to quantum networks for entanglement distribution due
For quick reference, classical routers are capable of finding optimal to their particularities compared to the transmission of classical bits,
routes in a complex network and understand the Internet Protocol such as:

7
D. Barral et al. Computer Science Review 57 (2025) 100747

• The duration of entanglement and the lifetime of the qubits due to layer architecture relies on the introduction of the connectivity layer,
decoherence. which is responsible for allowing point-to-point or point-to-multipoint
• The probabilistic nature of some quantum mechanisms. connectivity, as well as error correction and establishment of long-
• The need for mechanisms to improve fidelity, such as distillation. distance links. The link layer allows the creation of graph states in
• The possibility of joining entanglement links not only through sequen- the network that clients will subsequently use for the creation of
tial operations but also through operations carried out in parallel on end-to-end graph states.
the various links.
• The different entangled resources – bipartite, multipartite by means In relation to the protocols necessary for the QNs they are classified
of GHZ, W, cluster states, etc. in layers of the Open Systems Interconnection (OSI) model, in an
• The need for both quantum and classical channels to achieve the analogue way to the classical counterparts. Several implementations
desired functionality. of some of the functionalities of each communication layer have been
• The possible use of quantum networks not only for the transmission proposed:
of quantum information but also for the distribution of entanglement
between distant points, which can be used as a resource by itself. • Layer 1 (Physical): A. Dahlberg et al. define a protocol for physical en-
• The resource reservation strategy, if needed. tanglement generation based initially on NV center platforms [173].
A. S. Cacciapuoti et al. compare the addressing needs in quantum
Software development requires the definition of a set of protocols
networks and classical networks (for instance, entanglement to a
that satisfies the different communication requirements from the most
destination might require to perform entanglement to an intermediate
basic physical level communication to the application level commu-
nication, usually defined as a stack of network software layers. The node instead of the destination), and also the implications in the su-
survey [164] summarizes the main works on network protocol stacks, perposition of paths for both the addressing and routing design [174],
compared to the classical OSI or TCP/IP, and provides a comparison of while J. Miguel-Ramiro et al. assign each network device an identifi-
the different stacks. Also noteworthy is the publication of the Internet cation register and an activation register which depend on the target
Research Task Force (IRTF) Architectural Principles for a Quantum node [163]. When a Toffoli operation on both registers is successful
Internet [165], where the general guidelines for the design of quantum the node takes an active role in the operations.
networks are presented. Regarding hardware architecture, much work • Layer 2 (Link): J. Illiano et al. propose an entanglement access control
has focused on tackling the delivery of entanglement in different tech- mechanism (analogue to the ethernet MAC mechanism) to grant
nologies for the development of quantum repeaters. However, this is access to a subset of the nodes to the entangled state (contention
also to applicable to the development of quantum network devices for resource) [175]. The mechanism is based on quantum communica-
DQC: all optical and matter qubits architectures, based on discrete vari- tions using multipartite entanglement Dicke states and preserves the
able, based on continuous variable, based on bipartite entanglement or anonymity of the granted nodes. R. Hanson and S. Wehner define a
multipartite. In what follows, a list of relevant work related to DQC link layer protocol for robust entanglement generation sensitive to
about quantum network hardware architecture and software stacks is specific application needs (create and keep or create and measure,
presented:
number of entangled links, atomicity of the links creation, fidelity,
• [166] presents a CV all-photonic switch for entanglement creation and other relevant parameters for the link creation) [173].
among end-nodes that uses Gottesman–Kitaev–Preskill qubit encoding • Layer 3 (Routing): M. Caleffi designs a routing protocol and metric for
and Steane codes error correction. quantum networks considering the key parameters for entanglement
• [167] proposes an architecture for a CV continuous variable quantum generation and the needed optimization to determine the optimum
switch where end nodes share entanglement links error corrected by path between two points in the network [176].
means of Noiseless Linear Amplification (NLA) [168]. • Layer 4 (Transport): Yu et al. propose a protocol for the reliable
• Dür et al. [169] propose an architecture and network stack for transmission of quantum information [177]. The protocol is based
quantum networks based on multipartite entanglement (GHZ graph on the three way handshake of the classical counterpart TCP [178]
states) allowing the generation of graph states of any type among and on a recursive quantum secret sharing method ((2,3) threshold
clients. scheme [179]) to achieve the transmission of the quantum data
• Van Meter et al. [170,171] propose a Quantum Recursive Network reliably, where the message to be transmitted is encoded in segments,
Architecture (QRNA) describing the layers of network communica- being able to recover the message when only 2 out of 3 segments are
tions that tackle entanglement distribution end to end. They introduce available. If one of the segments is lost, one of the remaining segments
a recursive layer architecture in which swapping and purification will be reencoded and the method repeated.
functions are repeated to build end-to-end entanglement paths from • Layer 5 (Application)3 : T. Satoh, R. van Meter et al. include the
a sequence of links, being entanglement performed at link level.
design of quantum sockets in analogy to the classical communications
Physical and link layer are in charge of entanglement establishment
sockets, allowing the applications to access the services and having
at link level (point-to-point), while Remote State Composition and
similar functions (creation and destruction of the socket, connection
Error Management layers are recursive and are continuously repeated
to the socket; reading from the socket, writing to the socket and
performing swapping and purification from entangled links until the
system is able to build an end-to-end entangled path. configuring the socket) [171,181].
• Li et al. [172] and Dahlberg et al. [173] propose protocol architec-
Regarding the control plane, [182] propose a control architecture
tures for quantum networks based on bipartite entanglement where
for entanglement generation in quantum networks that moderates the
the mission of physical and link layers is the establishment of reliable
requests for entanglement resources with the goal of fair distribution
entanglement, the network layer’s goal is the establishment of long
distance entanglement, and the transport layer copes with the qubits of the service to the network end nodes. [183] proposes a protocol to
reliable/deterministic qubits transmission. manage entanglement requests and resources (memories) management.
• Pirker and Dür [169] propose an architecture and network stack for
quantum networks based on multipartite entanglement (GHZ graph
states) allowing the generation of graph states of any type among 3
A quantum operating system for quantum network applications has been
clients. This architecture is composed of four layers: physical, connec- implemented and experimentally tested running a client–server application on
tivity, link, and network. The main difference to the traditional OSI NV nodes in [180].

8
D. Barral et al. Computer Science Review 57 (2025) 100747

Fig. 4. Sequential phases of classic compiler process: analysis and synthesis stages.

4. Development layer

In the realm of classical computing, compilation serves two primary

purposes: translating complex programming constructs into machine-
specific executable instructions and optimizing machine resources to
produce efficient code. Typically, this process follows a common
scheme, as illustrated in Fig. 4, which consists of two main phases: anal-
ysis and synthesis. The analysis phase is responsible for conducting the
code’s lexical, syntactic, and semantic analysis to ensure correctness.
Once validated, the code is translated into an Intermediate Representa- Fig. 5. Types of quantum distribution and their stages simplified.
tion (IR), which simplifies the implementation of optimizations in the
synthesis phase.
Regarding quantum compilation, the scheme followed is usually
and managing the quantum circuit across different QPUs. For instance,
the same as in the classical world. This is mostly because quantum
Fig. 6 illustrates how Bell pairs can be created in the three main
compilation turns out to be a fully classical task, leaving the quantum
distribution categories. The following sections will dissect each of these
workload just for the execution part. This leads to the situation where
categories to fully understand how the quantum distribution works in
many quantum development software tools are actually built on top of
each case.
classical languages, allowing the analysis phase to be integrated into
an existing implementation.
Adding distribution to this task does not alter the compilation 4.1.1. Circuit distribution
scheme; it remains largely the same with some additional steps and Circuit distribution, as shown in Fig. 5, involves three main phases:
restrictions. To fully picture the differences and intricacies of compiling first, finding an optimal or near-optimal partition; second, distributing
a distributed program, this section will be divided into two parts: Sec- the partition among the available QPUs, and third, mapping this parti-
tion 4.1 will elucidate the various methods by which a quantum process tion to each QPU. However, partitioning the circuit presents the most
– usually referred to as a quantum circuit – can be distributed, while significant challenge and will be the primary focus of our efforts in this
Section 4.2 will delve into how the compilation process is executed section. The other aspects are common to all the distribution types and
considering the distributed nature of the task. will be further explained in the compilation Section 4.2.
First, for partitioning, the quantum circuit is mapped onto a graph
4.1. Types of distribution that shows interconnections between elements. Thus, quantum circuit
partitioning turns into a graph partitioning problem: given an undi-
Distributed computing makes it possible to organize the compu- rected graph 𝐺 = (𝑉 , 𝐸) with a vertex set 𝑉 and an edge set 𝐸, the aim
tation of a problem in different Processing Units (PUs), which are is to partition 𝑉 into two or more subsets regarding a cost function,
connected through an interconnection network. The advantages of like the number of edge cuts generated by the partition.
this model are evident: reducing the execution time by leveraging Graphs assume that the interaction between vertices is by pairs.
multiple PUs computing in parallel or, for large problems that do not However, even the most trivial phenomenon implies more than two
fit within a single node, partitioning them to enable their solution. The vertices interacting concurrently. It is necessary to broaden the graph
time reduction comes with its own set of disadvantages, notably the concept to gather these multilateral connections. The so-called hyper-
increased difficulty in adapting algorithms and codes to a distributed graphs [186] generalize the graphs to more complex situations. In short,
approach. This is due to the significant overhead caused by communi- while a graph can establish connections by pairs, a hypergraph is an
cations and synchronizations, which must be carefully considered and object that connects more than two vertices or pins through elements
managed [184]. called hyperedges or nets, as shown in Fig. 7. Thus, a hypergraph
Therefore, the complexity of developing a code increases when it 𝐻 = (𝑉 , 𝐸) is an ensemble of pins 𝑉 and nets 𝐸 among those pins,
is distributed. This complexity especially impacts the compiler design. and a net 𝑒 ∈ 𝐸 is a subset of more than two pins.
In the analysis phase, new communication directives need to be devel- Hence, hypergraph partitioning generalizes graph partitioning.
oped, while in the synthesis phase, various network architectures must More precisely, a 𝑘-way hypergraph partitioning groups the pins of a
be considered to optimize data transmission and reception [185]. hypergraph into 𝑘 blocks minimizing an objective function so that few
Certainly, the network’s communication mechanisms and the re- nets connect pins from different blocks. The exchangeable objective
sources the quantum task requires dictate the applicable distribution functions are the cut-net and the connectivity metrics. The cut-net
model, as depicted in Fig. 5. Three distinct categories of quantum metric generates independent blocks of vertex sets by minimizing
distribution emerge: circuit distribution, circuit cutting, and embarrassingly the nets belonging to several blocks, whereas the connectivity metric
parallel. It is clear, looking at Fig. 5, that all categories converge in com- weights each net 𝑒 with a factor 𝜆𝑒 − 1 to diminish the 𝜆𝑒 blocks
piling, executing, measuring, and post-processing information. Each of connected by a net. The cut-net objective function sums over the nets
these distribution methods involves specific strategies for partitioning among blocks and the connectivity metric over the 𝜆𝑒 blocks connected

9
D. Barral et al. Computer Science Review 57 (2025) 100747

Fig. 6. Example of the creation of a Bell pair for each type of distribution studied. In all types a partition of the tasks or the circuit (dotted line) is specified.

options for executing global gates, which limits their ability to produce
optimal solutions.
These limitations were pointed out in the work of Houshmand
et al. [192], who improved on the work of Zomorodi et al. by exchang-
ing the algorithm responsible for reducing teleportations – which had
exponential cost – for a genetic one, which allowed them to signifi-
cantly reduce the execution time. This work, like the Zomorodi et al.
one, only considers a two QPU scheme, reason why Daei et al. [193] en-
hanced it by effectively mapping a quantum circuit into an appropriate
number of distributed components. Moreover, Nikahd et al. [194] also
go beyond, categorizing the binary gates into distinct ‘‘levels’’, followed
by determining the optimal partitioning of qubits for each level through
Fig. 7. Example of a hypergraph with twelve pins 𝑣𝑖 and four nets 𝑒𝑗 . Net 𝑒1 has a the solution of an integer linear program.
size of four as it ensembles four pins, and pin 𝑣4 has a degree of 2 as it belongs to The work by Martínez and Heunen [189], on the other hand,
two nets. was extended with an entanglement-efficient protocol [195] derived
from [15] and with, among other things, a hypergraph approach to
arbitrary network topologies [196]. In the first case, authors pack
by a net. Nevertheless, both are analogue to the edge-cut problem in multiple non-local controlled unitary gates locally with one maximally
graph partitioning. entangled pair through a distributing and embedding pipeline. In the
Underneath the goal of minimizing the cut-net and connectivity second, the authors also search for efficient entanglement within the
metrics lies an important consideration: while a valid partition may network by reusing already available connections. In fact, this work
suffice for DQC, it may not necessarily be an optimal partition. For led to many different articles employing hypergraph partitioning with
instance, in the circuits responsible for teledata and telegate operations KaHyPar.
– as illustrated in Fig. 3 –, these operations add up to four layers of Following the KaHyPar line, Sundaram et al. contribution con-
depth to the circuit to enable operations among qubits in different cerns communication timing, non-local operations availability (teledata
QPUs. Consequently, this introduces latency to the quantum circuit, and/or telegate), and partitioning. First, engaging KaHyPar, Sundaram
especially considering the additional synchronization required for in- et al. [197] present a two-step heuristic for the distribution of quantum
termediate measurements contained in both protocols between both circuits: dividing the given qubits among the computers in the network
QPUs. This latency represents a significant bottleneck in circuit dis- with KaHyPar and scheduling communication operations, called migra-
tribution. Therefore, all circuit partitioning methods aim to minimize tions – equivalent to cat-entanglement operations [53]. They present a
the utilization of teledata or telegate protocols. This aspect will be polynomial-time solution for the second step in a specific setting and
crucial in the circuit distribution techniques discussed in this section a O(log 𝑛)-approximate solution in the general setting. Second, they
and beyond, summarized in Fig. 8. amplify the available remote protocol for communications between
Zomorodi et al. [187] introduced a first approach aiming to reduce QPUs, [198]: while Daei et al. [193] use teledata and, on the contrary,
communication between partitions, considering only two QPUs. They Martinez and Heunen [189] and Sundaram et al. [197] use telegate,
use the Kernighan-Lin (KL) [188] algorithm, a heuristic algorithm there is upgrading in Sundaram et al. [198] applying both. For the
for graph partitioning, to divide the graph vertices into two subsets telegate protocol, they consider a method similar to the initial two-
to reduce the edges across the subsets to minimize communication step heuristic work, [197]. However, to partition the given qubits
between the two partitions. After that, they apply a custom algorithm among QPUs, they use a Tabu-search-based heuristic regarding the
that aims to reduce the number of teleportations applied. heterogeneity of the network and storage limits. For the general DQC
The work of Martínez and Heunen [189] was one of the most problem, they employ two heuristics: Sequence, a greedy approach, and
significant contributions in the field, serving as a foundational reference Split, similar to the previous one, but with an iterative approach. Both
in many of the articles discussed here. Their method involves two contemplate the telegate solution as a subroutine. Lastly, Sundaram
key phases: a pre-processing phase, which groups equivalent gates, et al. take a step further in a recent work [199] by designing two
and a second phase, where hypergraph partitioning is performed using different protocols to reduce the number of teleportations needed to
Karlsruhe Hypergraph Partitioning (KaHyPar) [190,191], a multilevel perform the distributed task. The first method, termed Local-Best, tries
hypergraph partitioning framework that enhances cut-net and con- to minimize the teleportation of qubits by selecting them only when
nectivity metrics. They evaluated their algorithm using five quantum necessary, with the choice of teleportation influenced by gates in the
algorithms known for their quantum speedup, such as Quantum Fourier near future. The algorithm consists of two steps:
Transform (QFT). A criticism of this work is that it did not consider
optimizations such as moving gates back and forth to bring them closer 1. Find an initial assignment of qubits to computers to minimize the
together nor explore the entire search space of different partitioning number of resulting non-local binary gates.

10
D. Barral et al. Computer Science Review 57 (2025) 100747

Fig. 8. Scheme of contributions to quantum circuit partitioning.

2. For each non-local binary gate G, select the teleportations to execute from the previous time slice and utilizing a tunable lookahead scheme
G locally based on the ‘‘near future’’ in order to minimize the total to reduce the cost of moving to future time slices. To achieve this,
number of teleportations. a customized version of the Overall Extreme Exchange (OEE) algo-
rithm [205] – considered a natural extension of the KL algorithm –
The second method to shorten the number of teleportations, Zero- referred to as relaxed-OEE (rOEE), is employed. Because the primary
Stitching, comprises also two steps: approach to map the circuit to the hardware is Fine Grained Parti-
tioning (FGP), this method is usually referred to as FGP-rOEE. This
1. Identify ‘‘zero-cost’’ subcircuits: These are contiguous subcircuits
method was further analyzed by Ovide et al. examining it under an-
that can be executed without any teleportations.
other multi-core architecture but maintaining the all-to-all qubit and
2. Divide the given circuit into zero-cost subcircuits and ‘‘stitching’’
cores connectivity [206]. Moreover, a Hungarian Qubit Assignment
them together using teleportations.
(HQA) method for partitioning is developed by Escofet et al. which also
There are also approaches employing bipartite graphs instead of describes the assignment of qubits to cores between timeslices, and it
hypergraphs. Davarzani et al. [200] introduce an algorithm for dis- is compared to the FGP-rOEE method [207].
tributing quantum circuits to optimize the number of teleportations A recent approach that has elevated the work of Baker et al. is
between qubits that consists of two steps: first, the quantum circuit is the technique presented by Bandic et al. which relays on Quadratic
converted to a bipartite graph (bigraph), and, second, the bigraph is Unconstrained Binary Optimization (QUBO) to partition the circuit at
partitioned into 𝐾 parts employing a dynamic programming approach. each time slice [208]. Their method’s primary strengths are rooted in
Finally, they compare their results with the ones yielded by works pre- the formulation of the QUBO itself. This structure enables the decou-
viously analyzed [187,189,192] and they claimed that the experiments pling of the problem definition from the solver as well as surpassing
gave better or equal results for benchmark circuits. the limitations of look-ahead approaches utilized in the Baker et al.
In another approach, proposed by Clark et al. [201], a different solution. It is worth noting that, in this approach, two different multi-
model than hypergraph is employed. They introduce the Tree-based core architecture layouts composed of 10 cores with a capacity of 10
Directed Acyclic Graph (TDAG) partitioning for quantum circuits, a qubits each were tested, in contrast with the non-realistic all-to-all
novel method that views circuits as a series of binary trees and selects connectivity assumed by the previous approaches.
the tree containing the most gates for partitioning. Last but not least, one of the most novel algorithms is a cir-
Besides minimizing the communication between partitions, Cambi- cuit partitioning method that employs Deep Reinforcement Learning
ucci et al. suggest [202] adjustable scenarios to the capabilities and (DRL) [209]. Once again, the FGP-rOEE is employed as a baseline
constraints of the processing units involved in the distribution are con- to compare the results and as an inspiration due to its time-sliced
sidered. In this work, instead of the KL from the original hypergraphic graph partitioning. This work has considered three approaches: Proxi-
approach, authors implement a variation of the Fiduccia–Mattheyses mal Policy Optimization (PPO), Soft Mask, and Hard Mask. The first
algorithm [203], which is a faster approximation algorithm for min- one, the PPO, is a widely used algorithm within the DRL scheme,
cut partitioning with a computational time that grows linearly with the while the remaining two, Soft and Hard Mask, are a variant of the
network size. They use the same circuits as [189] for benchmarking. former PPO algorithm that introduces a masking mechanism. The Soft
A field-changing approach was the work developed by Baker et al. Mask approach adds a simple mask, which disables useless operations
[204]. While still based on graph partitioning, this method seeks to – such as swapping identical qubits, swapping two qubits situated on
avoid reaching a single static assignment for an entire circuit by em- the same machine, or advancing to the subsequent time slice without
ploying near-optimal graph partitioning techniques. It leverages the establishing a valid assignment for the current one – whereas Hard
inherent clustering of the DQC paradigm and the statically-known Mask implements a direct-swap heuristic in top of the Soft Mask which
control flow of quantum programs to develop tractable partitioning solely evaluates the relocation of misplaced qubits to the respective
heuristics. These heuristics map quantum circuits to modular physical core they need to interact with.
machines one time slice at a time. Specifically, optimized mappings Now that we have explored the state-of-the-art in the circuit par-
are created for each time slice, considering the cost to move data titioning problem, we can understand why it poses such a significant

11
D. Barral et al. Computer Science Review 57 (2025) 100747

challenge. Finding the optimal partition directly impacts performance

and is a critical aspect in the later stages of compilation, where the
boundaries between software and hardware become narrow. Specifi-
cally, this problem is closely related to the qubit mapping and circuit
optimization stages of the distributed quantum compiler, which will
be defined and explained in Section 4.2.3 as part of the compilers’s
synthesis phase.

4.1.2. Circuit cutting

As detailed in Section 3, on the road to fully functional DQC, one
needs quantum communication in the form of a quantum network
between the devices. In the absence of such of networks, there are
several alternative techniques to simulate, or at least approximate, this
entanglement using a classical network. In this context, circuit cutting
has been suggested as a solution to partitioning a wide circuit requiring
many qubits into smaller, non-entangled subcircuits. These subcircuits
can then be executed (emulated) sequentially on a limited-qubit (mem-
ory) device or in parallel across multiple devices. There are several
different strategies for circuit cutting, such as gate-cutting and wire-
cutting (shown in Fig. 9), which will produce different subcircuits. The
output of the original circuit is recovered using a combination of the
results of the subcircuits, with some cost in accuracy that is overcome
by increasing the number of circuit executions. This extra cost is often
called sampling overhead, and it is known to grow exponentially with Fig. 9. Two schemes for cutting a quantum circuit: gate-cutting (or spatial cut)
the number of cuts. as shown by Mitarai and Fujii [223] and wire-cutting (or temporal cut) by Peng
et al. [224]. Both can be shown to be equivalent to simulating teleportation [225],
Quasi-probabilistic decomposition of quantum channels. Most circuit- wire-cutting being analogous to teledata and gate-cutting to telegate.
cutting algorithms rely on the quasi-probabilistic simulation (QPS) of
a quantum circuit, which uses the quasi-probabilistic decomposition
(QPD) of the quantum channel of the circuit. A quantum channel , or simulate quantum processes by separating the ‘‘hard’’ and the ‘‘easy’’
quantum operation, is a trace-preserving, completely positive linear map parts of the circuit [213,214], and also for performing error mitigation
between density operators. Quantum channels are typically represented through a quasi-probabilistic decomposition of an ideal circuit from
through the operator-sum representation, also known as Kraus decom- noisy ones [215,216].
position. In this representation, a channel  acts on a state described In practice, to calculate the expected value of an observable, we
∑
by a density matrix 𝜌 as a sum of 𝑘 terms (𝜌) = 𝑘𝑗=1 𝐸𝑗 𝜌𝐸𝑗† , where 𝐸𝑖 sample the outcome of the circuit measured in the appropriate basis
are (Kraus) operators on the Hilbert space of 𝜌. for some number of shots 𝑁𝑠 . We want 𝑁𝑠 to be large enough so
This representation is not unique, i.e., one has the freedom to choose as to have some desired degree of accuracy 𝜖. When using QPS to
the operators 𝐸𝑖 of the representation and still get the same channel . simulate circuits, the variance of the result increases with 𝜅 2 , and
In particular, one can choose the operators to be quantum gates that are we have to compensate for increasing 𝑁𝑠 proportionally. This effect
local in separate sets of qubits. Consider the 𝑛-qubit bipartite system is known as sampling overhead. This overhead is multiplicative, in-
𝜌 = 𝜌(1) ⊗ 𝜌(2) with Hilbert space  =  (1) ⊗  (2) , where  (1) and creasing exponentially with the number of cut gates 𝑁𝑐 . Given a
 (2) are the space of the two unconnected sets of qubits 𝜌(1) and 𝜌(2) . large enough number of shots, the outcome of the original circuit is
Now consider a quantum circuit 𝐶 consisting of products of arbitrary recovered with arbitrary precision. However, noise sources will still
quantum gates, some of them multi-qubit gates acting on both  (1) and introduce a bias in the computation independent of the QPS, as noise
 (2) simultaneously. Our hardware may not be able to execute those
is a separate quantum channel evolving the state 𝜌. Quasi-probabilistic
non-local gates, but one can always find a decomposition such that
methods can also aid in error mitigation, which as mentioned above
∑
𝑚 ( )( )( ) has some practical overlap with circuit cutting. Furthermore, there is
(𝜌) = 𝑞𝑖 𝑉𝑖(1) ⊗ 𝑉𝑖(2) 𝜌(1) ⊗ 𝜌(2) 𝑉𝑖(1)† ⊗ 𝑉𝑖(2)†
experimental evidence that QPS can reduce the effect of noise sources
𝑖
∑
𝑚 ( ) ( ) by employing smaller circuits [217,218]. Another issue appearing when
= 𝑞𝑖 𝑉𝑖(1) 𝜌(1) 𝑉𝑖(1)† ⊗ 𝑉𝑖(2) 𝜌(2) 𝑉𝑖(2)† sampling a QPS appears when reconstructing the evolved 𝜌 from the
𝑖 partitions. Due to finite sampling error, finding a distribution with
∑
𝑚
( ) ( ) negative terms is possible. To solve this, post-processing can be used
= 𝑞𝑖 𝑖(1) 𝜌(1) ⊗ 𝑖(2) 𝜌(2) ,
𝑖
to find the ‘‘most likely’’ output state [219,220], although this is not
∑ (1)
necessary for calculating expected values of observables.
with coefficients 𝑞𝑖 ∈ R with 𝑚 𝑖=1 𝑞𝑖 = 1, and 𝑉𝑖 and 𝑉𝑖(2) are oper- Finding an efficient QPD of a general circuit 𝐶, i.e., a QPD with
ations acting locally in  (1) and  (2) respectively, that our hardware a small 𝜅, is difficult. If the circuit is known to produce a state with
can physically execute. The choice of 𝑞𝑖 and the set of 𝑉𝑖(1) and 𝑉𝑖(2) is a particular bi-partite structure, one can turn to similar techniques to
not unique, and it is known as a QPD of the quantum channel [210]. execute the parts locally, such as Entanglement Forging [221,222].
The 𝑞𝑖 can be either positive or negative, which is why they are
However, the main direction that has been followed in the literature
called quasi-probabilities. The larger the number of negative coeffi-
∑𝑚 for circuit cutting was to perform only the QPD of specific regions of
cients in the decomposition, the larger the 1-norm 𝜅 = 𝑖=0 |𝑞𝑖 | of the circuit with sparse correlations, targeting non-local gates or wires.
the QPD becomes. Crucially, this 𝜅 quantity is related to the cost of
executing the circuit 𝐶 that has non-local gates, using only local opera- Circuit cutting techniques: gate-cutting and wire-cutting. One preliminary
tions [211,212]. Negative probabilities in the simulation of quantum work, which was later labeled as circuit cutting (and in particular, wire-
circuits were already known to be related to the ‘‘quantumness’’ of cutting), was the cluster simulation scheme by Peng et al. [224], which
quantum circuit. Thus they could be used as a resource to classically decomposes the corresponding tensor network of a given quantum

12
D. Barral et al. Computer Science Review 57 (2025) 100747

circuit into smaller clusters. Inter-cluster communication is then sim- et al. realize this by introducing pre- or post-selection methods for
ulated classically. The authors apply these techniques for Hamiltonian quantum channels [242], while Chen et al. use approximate methods
simulation using the Variational Quantum Eigensolver (VQE) [226], that directly neglect some of the elements [243,244].
and suggest using this hybrid variational ansatz for future modular Efforts to minimize quantum communication between machines fo-
architectures. Later, Mitarai and Fujii [223] introduce the idea of virtual cus on smart qubit assignment. A solution that minimizes the sampling
two-qubit gates, where the action of the virtual gate is substituted with overhead also minimizes the number of Bell pairs in a DQC protocol,
local operations. This way they only apply QPS for the non-local gates and thus, the same compiling tools could be used for both techniques.
we want to get rid of. Given that most QPUs can only execute single- Combining gate- and wire-cutting finds better partitions [245], which
and two-qubit gates, it is more convenient to find an efficient QPD of is crucial for DQC, not only for circuit cutting, as already detailed
the particular two-qubit gate and simulate them with local single-qubit in Section 4.1.1. Some Software Development Kits (SDKs), such as
gates. The total overhead of the QPS then scales as O(𝜅 2𝑁𝑐 ) with 𝑁𝑐 Qiskit or Pennylane, incorporate these techniques in their compilation
being the number of virtual gates. Mitarai and Fujii also provide an routines. Moreover, several tools such as CutQC [246], ScaleQC [247]
efficient QPD for a two-qubit gate with 𝜅 = 3 at most, from which or SuperSim [248] perform the whole circuit cutting pipeline, finding
most common two-qubit gates such as 𝐶𝑁𝑂𝑇 , 𝐶𝑍, 𝑅𝑍𝑍(𝜃), etc., can cuts, executing the subcircuit, and reconstructing the state. There is
be derived. Fig. 9 compares the two methods, which can also be used also, as we will delve in Section 4.2, a compiler named Qurzon [249]
simultaneously in the same circuit. which performs all the aforementioned techniques – in fact, it uses
The main drawback of circuit cutting is the exponential overhead. CutQC in combination with other tools.
This overhead has been proven to be strictly exponential [227], so Herzog et al. [250] illustrated the practical application of these
it cannot be reduced to a polynomial increase and it proves a big methods by cutting a QAOA ansatz for a combinatorial optimization
challenge for scaling to large problems. Still, minimizing 𝜅 is an active problem. Their approach combined the strategies in [224,232] to re-
research topic. In [225,228], the minimal sampling overhead to cut duce ancilla and classical communication requirements, while utilizing
wires and two-qubit gates is derived analytically. Brenner et al. [225] classical graph shrinking techniques to lower the overall overhead.
show that cutting an identity gate that transported the state of the qubit However, the authors noted that the same classical techniques could
before and after the cut (a wire cut) is equivalent to a teleportation potentially solve the problem faster through purely classical computa-
protocol. As shown in Section 2.2, one needs a prepared Bell state tion. Similarly, IBM’s recent work [229] demonstrated the execution of
and two bits of classical communication to teleport one qubit of data. a 142-qubit graph state across two 127-qubit QPUs. By implementing
Gate-cutting of a Bell pair between two qubits (𝜅 = 3) is already more Piveteau et al.’s concept of a Bell State factory, they utilized real-time
efficient than cutting a wire (𝜅 = 4), although it requires ancilla qubits. classical communication and parametric circuits to reduce compila-
Piveteau et al. [228] suggests that this overhead can be reduced tion time, showcasing the practical application of circuit cutting in
when jointly cutting multiple gates or wires, using classical commu- large-scale quantum computations.
nication between partitions. This is because the joint QPD of a larger Despite these advancements, circuit cutting occupies a challenging
unitary of 𝑁𝑐 Bell states (also called Bell State factory) required for position in the quantum computing landscape. It is more suited for
teleporting 𝑁𝑐 gates has a lower overhead than individually cutting 𝑁𝑐 problems with sparse entanglement, which are often more easily tack-
Bell states. This overhead now scales better 𝜅 = (2𝑁𝑐 +1 − 1), albeit using led using classical methods. At the very least, circuit cutting can be
local operations and classical communication (LOCC) and ancilla qubits useful for early DQC applications, serving as a transitional strategy
as requirements (one per partition and cut). While Piveteau et al. did until robust quantum communication networks are fully realized. This
not give the explicit QPD of this Bell State factory, it was later provided technique could thus provide a crucial stepping stone in the evolution
in [229] for 𝑁𝑐 = 2 and 𝑁𝑐 = 3. of quantum computing infrastructure.
Lowe et al. [230] reduced the ancilla qubit requirements for large-
scale Quantum Approximate Optimization Algorithm (QAOA) simu- 4.1.3. Embarrassingly parallel
lations by combining wire-cutting and random measurement bases The term embarrassingly parallel was coined within the HPC domain
(inspired by classical shadow tomography [231]) and subsequent stud- to describe applications that are inherently amenable to paralleliza-
ies improved bounds for multiple wire-cuts with LOCC [232,233]. tion without significant effort. Notable examples include bag-of-tasks
For gate-cutting [234] improved on Piveteau et al. result, finding an workloads – jobs devoid of dependencies that can be executed in any
optimal decomposition of an arbitrary two-qubit rotation gate and sequence – and parameter sweep applications, which involve numerous
reducing the ancilla requirements for cutting multiple parallel gates. parallel executions with varying parameter configurations.
Soon after, [235] achieved a similar result for clustered Hamiltonian Similarly, in the context of quantum computing, the term embar-
simulation, and [236] did the same for general two-qubit unitaries. rassingly parallel refers to the scenario where a problem can be divided
Reducing sampling overhead can also be achieved by cutting larger into multiple smaller computations that can be executed independently
unitaries. For instance, cutting a SWAP gate using QPS results in a without the need for direct communication among them. The simplest
lower overhead (𝜅 = 7) than decomposing it into three CNOT gates example of this in the quantum case is the distribution of shots, where a
and cutting each individually (𝜅 = 33 ). This, of course, is the idea quantum algorithm or kernel needs to be executed multiple times with-
behind cutting the Bell State factory in [228] but can also be extended out any structural changes – except for the modification needed to map
to higher dimension operators like Toffoli gates [236], multi-controlled the circuit to the different QPUs. Despite the quantum nature of the
CZ gates [237], and even the QFT [238]. Furthermore, in the case tasks involved, this method essentially involves classical parallelism.
of Variational Quantum Algorithm (VQA) one can choose variational A different approach comes from a distribution of the circuits
ansatzes designed with reduced entanglement between parts, so they needed to reconstruct the expectation value of a given observable or
are easier to partition. This can be in the form of clustered ansatzes to support the optimization protocol. This allows several possibilities:
for VQE [239,240], or more general ansatzes where the amount of
entanglement is tuned so that the overhead of the QPD is always kept • Distribution of terms in an observable. The distribution of the ex-
∑
below a tolerance value [241]. pectation value terms ⟨𝑂𝑖 ⟩ of a given observable ⟨𝑂⟩ = ⟨𝑂𝑖 ⟩
Other strategies reduce the number of subcircuits in decompositions is a case of embarrassingly parallelization. An intuitive example
to lower sampling overhead. Note that, while related in their expo- is the VQE [226], where the function to minimize is the energy,
nential scaling, the number of subcircuits in a QPD (its 0-norm) is not i.e., the expectation value of a Hamiltonian ⟨𝐻⟩. Depending on the
the same as the sampling overhead (its 1-norm). Reducing subcircuits specific problem, Hamiltonians can be commonly expressed using
can aid scheduling and post-processing without increasing 𝜅. Nagai fermionic operators in second quantization formalism, as in the case

13
D. Barral et al. Computer Science Review 57 (2025) 100747

of many systems in condensed matter/chemistry, bosonic operators, qubit mapping, device benchmarking, crosstalk4 characterization, or
or directly in Pauli operators, as in spin Hamiltonians that apply even vulnerability analysis [266–273].
to different problems in physics, route optimization, protein fold- Another paradigm that may be interesting to delve into is quantum
ing [251], and scheduling, among others. In all cases, except the last offloading. As mentioned in the introduction, QPUs is intended to be
one, the Hamiltonian has to be mapped to qubit instructions via some seamlessly integrated into classical HPC infrastructures, working along
encoding techniques [252,253]. After that, it appears as a weighted other hardware accelerators. This way of distributing the workload
sum of tensor products of Pauli operators, most commonly known as allows concurrent computations of classical and quantum tasks, let-
Pauli strings. Initially, each Pauli string can be individually sent to ting CPUs proceed with calculations while QPUs accelerate specific
different QPUs. However, the scaling in the number of Pauli strings processes in which the so-called quantum advantage takes part.
for complex problems makes this procedure inefficient. A common A profound quantum offloading analysis diverges from this work’s
practice is to form groups of Pauli strings that will share the same main scope, but some relevant works can be outlined. For instance,
quantum circuit to construct their expectation value. These groups are the eXtreme-scale Accelerator programming framework (XACC) is a
made of commuting Pauli operators that are determined using some system-level software infrastructure for quantum-classical computing
classical routine. The simplest strategy is qubit-wise commutativity, that promotes a service-oriented architecture to expose interfaces for
where each of the commuting groups built can be measured using core quantum programming, compilation, and execution tasks [8].
a single quantum circuit without difficulties [252]. An alternative is Strongly related is QCOR, a language extension specification of C++
general commutativity, which is more efficient in reducing the number that enables single-source quantum-classical programming and that
of commuting groups but entails the non-trivial task of finding the employs XACC as a base [9]. Another work leveraged the OpenMP
appropriate unitaries for the joint measurement of the groups [252, API to target quantum devices, which provides an easy-to-use and
254]. efficient interface for HPC applications to utilize quantum computing
• Gradient and Hessian’s distribution. Just like the preparation of a resources [274]. Similar to this were the efforts made to add QPUs to
parameterized trial wave function |𝜓(𝜃)⟩ to our problem, first and the OpenCL ecosystem of execution [7]. Even the NVIDIA company
second partial derivatives of the state |𝜓(𝜃)⟩ can be analyzed with a has developed the CUDA Quantum Platform for hybrid quantum–
quantum computer [255–257]. In many cases, the quantum circuits classical computation [275], enabling the aforementioned integration
that arise from the partial derivatives can be expressed as a linear and programming of QPUs along with other accelerators.
combination of circuits that use the same structure of the original
circuit to prepare |𝜓(𝜃)⟩, with a shift in their parameters, which is 4.2. Compilation
known as parameter shift rule [258].
• Distribution in a gradient-free optimization. That is a particular case of After resolving the distribution challenge, it is essential to explore
distribution that sources from the usage of gradient-free optimizers the compilation process thoroughly. We will adhere to a structure
such as evolutionary ones. These optimizers overcome the need to akin to the classical approach, which involves an analysis phase, an
compute gradients at the cost of using several individuals/particles intermediate representation referred to as Quantum Intermediate Rep-
that interact in a certain way to modify their parameters or generate resentation (QIR), and a synthesis phase. This framework will aid
other candidates. That is the case, for example, of Differential Evolu- in comprehending the compilation process for DQC and underscore
tion and the Particle Swarm Optimization algorithms [251,259,260]. the disparities between classical and quantum computing in terms of
Each individual is a different set of parameters that can be executed in compilation.
parallel using the same quantum circuit structure. One of the possible
benefits of the previously mentioned optimizers is that they can 4.2.1. Analysis phase
mitigate problems in the optimization landscape [259,261]. However, The analysis phase in the distributed and monolithic quantum com-
this would come at the cost of increasing drastically the number of pilation is quite similar, with the additional challenge in the distributed
circuit executions. case of limited literature and software development compared to the
• Distribution of data. As in the case of classical Machine Learning, monolithic counterpart. In the monolithic scenario, the underutilization
another possibility is to distribute the data or the model during the of standalone languages is not because they do not exist; rather, options
training. For example, [262] proposes a tool for distributing training like Scaffold [276], Q# [277], isQ [278], Q|𝑆𝐼⟩ [279], among others,
of Quantum Machine Learning models that can also be used for VQEs. are available. However, they are less favored due to the need for
A federated approach has also been proposed [263]. users to understand and adapt to these languages. In contrast, libraries
like Qiskit [280], Cirq [281] and Qulacs [282], built on well-known
There are some packages that permit the distribution of these kinds classical languages such as Python (Qiskit and Cirq) and C++ (Qulacs),
of jobs among several QPUs [262,264], based on a master-worker are more widely adopted. This situation is even more pronounced in
architecture. These packages must cope with additional issues not seen the distributed case because there is a shortage of standalone languages
in classical Machine Learning distributed learning, such as the different specifically designed for distributed purposes. Consequently, the previ-
architectures of the QPUs (different gate sets, different topology, or ously mentioned quantum monolithic libraries are often repurposed to
different timing for execution), the noise of each single QPU and the simulate the distributed structure.
possible drift of these errors with the time, for counting some of the This is the case for Quantum MPI (QMPI) [283], which represents
current challenges. Additionally, these techniques can also be used an extension of the Message Passing Interface (MPI) protocol for dis-
when circuit cutting is applied. tributed quantum systems. We refer to this as a formal approach due
Another paradigm that can be considered in this context is multi- to the absence of a usable library that allows for actual or simulated
programming of quantum computers. The segmentation of a QPU, better DQC. However, a reference implementation for QMPI has recently been
known as multi-programming in quantum computing, can maximize published [284], although none of the code is available for use, neither
the hardware throughput – the number of used qubits divided by in open source nor as a binary, to the best of our knowledge.
the total number of qubits – and reduce the runtime. The pioneering The aim of QMPI is, obviously, to add quantum functionalities to
work for multi-programming by Das et al. [265] advocated for its an already widely used specification such as MPI. For this purpose, it
use to enhance the utilization and throughput of Noisy Intermediate-
Scale Quantum (NISQ) computers, wherein the qubits are employed
to execute multiple workloads concurrently. Other works introduce en- 4
Crosstalk is an unwanted coupling between qubits. It is one of the noise
hancements like selecting the appropriate number of circuits to execute, sources in NISQ devices and can condition the hardware throughput.

14
D. Barral et al. Computer Science Review 57 (2025) 100747

Fig. 10. The significance of intermediate representation in the compilation process -

Facilitating decoupling between high-level and machine code.

defines two types of nodes: classical and quantum. The only difference
between them is that classical nodes cannot be the target of quantum
directives, whereas quantum nodes can manage both quantum and
classical calls. The core of this difference lies in the inherent distinc-
tion between classical datatypes and quantum datatypes – bits and
qubits – along with the inclusion of EPR pairs, a crucial element for
the development of quantum communication protocols, as shown in
Section 2. Other than that, although MPI is much more advanced than
QMPI, as expected, the communication modes supported by the latter
are the same: point-to-point communication and collective operations.
Moreover, they define a simple performance model called SENDQ. It is
worth mentioning that, contrary to almost all literature on DQC, they
anticipate a relatively low logical clock speed for quantum computers
due to the overhead introduced by the quantum error correction.
Consequently, they do not expect classical communication to signifi-
cantly affect performance, choosing to ignore classical communication
in the SENDQ model. This approach contrasts significantly with all the
circuit distribution methodologies discussed in Section 4.1.1, where
Fig. 11. InQuIR representation of the creation of an EPR pair using remote gates.
the focus was primarily on minimizing the number of teledata and
telegates, considered the main bottleneck of quantum distribution. The
SENDQ model is closely associated with the NISQ era and may not be
sustainable when transitioning to the fault-tolerant era. node 1. InQuIR automatically adds the necessary directives to do the
Anyway, as it is explained in Wakizaka [285], there is a need to remote operation using the telegate technique.
develop a proper quantum programming language that takes consider- The IR code extends the basic quantum operations to a distributed
ation of a distributed structure and extracts profit from that structure setting, where quantum communication and entanglement generation
via advanced distributed computational techniques, just as it happens across different nodes (0 and 1) are involved. Lines 2 to 4 in both Figs.
in classical computation. 11(b) and 11(c) correspond to the initialization of the communication
channel between both nodes, the initialization of the local qubits, and
the generation of the EPR pair, respectively. Lines 5–6 in 11(b) and
4.2.2. Distributed quantum intermediate representation
5–7 in 11(c) correspond to the gates and measurements. The measure-
The compilation process is complex; therefore, Intermediate Rep-
ment results are transferred between the two nodes by send/recv
resentations (IR) were introduced to establish a break in the compiler
operations and used in the conditional gates.
in order to obtain modularity and decoupling [286]. An IR allows to
intermediate between the front-end and the back-end, improving the
4.2.3. Synthesis phase
efficiency of compiler development and allowing abstract optimizations
In classical compilation, this corresponds to the lowest level of
to the target machine. Fig. 10 shows the use of IRs as a break in abstraction. At this stage, low-level, less human-readable languages—
the compilation process to facilitate compiler development so that analogous to classical assembly languages—are utilized within the
programs are implemented for abstract machine code such as an IR. compilation chain. Although it is challenging to map each quantum
An important feature of IRs is that they have to be able to represent compilation stage to distinct levels of abstraction, a parallel with
the operations of different high-level languages to be implemented in classical assembly can be established through the use of Quantum
different machine codes. Therefore, with the evolution of quantum Assembly Language (QASM). There are a lot of different versions, such
computing, it is necessary to extend classical IRs (or create new ones) to as OpenQASM [294], cQASM [295], eQASM [296] and f-QASM [297].
include quantum instructions. This process has been evolving in recent But, to the best of our knowledge, only NetQASM [298] takes into
years, where the number of quantum IRs has grown considerably [287– account an underlying distributed structure.
292]. In [298], Dahlberg et al. introduced an abstract model featuring
For DQC, specialized IR are needed to allow the use of classical a Quantum Network Processing Unit (QNPU) for end-nodes in a QN.
and quantum communication instructions between different PUs. This NetQASM is proposed as an Instruction Set Architecture (ISA) designed
objective is what InQuIR [293], an IR specialized in DQC, aims to solve. to execute arbitrary programs on end nodes equipped with the QNPU.
To exemplify the operation of this IR, we use the circuit shown So, NetQASM can be seen as a low-level, assembly-like language tai-
in Fig. 3(b), which implements a CNOT remote gate between two lored for the quantum segments of quantum network program code.
separate nodes, but connected through a Bell pair |𝛷+ ⟩. Fig. 11(a) It specifies the interaction between the QNPU and executes QN code,
shows the OpenQASM code to implement this, which does not consider a functionality not available in other QASM languages. The language
communication directives. The compilation of OpenQASM to InQuIR is designed to be extensible, with a core set of instructions for clas-
produces the code shown in Fig. 11(b) for node 0 and Fig. 11(c) for sical control and memory operations and a set of quantum-specific

15
D. Barral et al. Computer Science Review 57 (2025) 100747

instructions grouped into ‘‘flavors’’. A ‘‘vanilla’’ flavor is introduced Regarding circuit cutting, optimizations aim at reducing the sam-
for universal, platform-independent quantum gates, enabling platform- pling overhead, or the number of subcircuits. Although both quantities
independent quantum network program descriptions, with the possi- are related in that both increase exponentially with the number of
bility of developing platform-specific flavors for optimized quantum shots, in general, they do not need to scale the same way. The most
operations on specific hardware. important of the two is the sampling overhead. Still, a reduction of
It is also worth mentioning the work of Ying and Feng [299]. the number of subcircuits (without an increase in the sampling over-
They developed an algebraic language for formally specifying quan- head) can also help in the scheduling and post-processing part of the
tum circuits in DQC that aims to represent circuits conveniently and computation. Some works reduce the sampling overhead by including
compactly, akin to how Boolean expressions are used for classical LOCC, either when jointly cutting several gates [301], or in smart
circuits. prepare-and-measure protocols in wire-cutting [230,232,233]. Other
Building on the classical analogy, this stage involves optimizing works attempt to cut larger unitaries [237] or constrain the overhead
the code and adapting it to the target machine. The compiler per- using parameterized gates [241]. Regarding the number of subcircuits,
forms operations such as register allocation, branch optimization, loop they can be reduced using pre- or post-selection methods [242], and
unrolling, and other well-known optimization techniques. Similarly, some of them can be neglected in approximated methods without
quantum compilation employs analogous optimization methods. How- incurring in large errors [243,244].
ever, unlike in classical compilation, these techniques are not always
applied directly to QASM. Instead, they can be applied to the higher- Qubit mapping. When it comes to classic computing, register allocation
level languages considered in this work. This distinction underscores is about finding the best way to use the limited number of registers
the current lack of abstraction in quantum computing. available to store variables [302]. In the field of quantum computing,
To maintain consistency with classical methodologies, the remain- qubit mapping can be compared to register allocation in classical
der of this section will elaborate on the three primary components of computing. This process involves finding an optimal mapping of logical
the synthesis phase: optimization, qubit mapping, and verification. First, qubits to physical qubits in a quantum device, taking into account
optimization and qubit mapping will be discussed, as they are funda- the device’s connectivity and other constraints. It is important to note
mental aspects of quantum compilation, particularly in the current the growth in complexity of this process as it moves from classical to
NISQ era. Finally, the verification stage will be examined. Although quantum compilation. In the realm of quantum compilation, it is not
verification differs in nature from the preceding two components, it only the use of the qubit’s value that must be evaluated – meaning
serves as a crucial feature in quantum programming by providing an if it is thought to be a communication qubit or a computing qubit.
alternative to classical debugging techniques, ans so it will be explained Other factors, such as the error associated with the specific qubit and its
at the end of the section as an important side aspect of the quantum interconnection with the remaining qubits, assume significance in the
compilation. decision-making process. Qubit mapping is an NP-hard problem [303].
Optimization. The optimization phase in monolithic quantum computing Therefore, exact algorithms are only computable for a reduced number
encompasses a broad range of techniques aimed at minimizing various of qubits, making it necessary to use techniques that are able to obtain
metrics, such as the number of 2-qubit gates, the circuit depth, etc. In an optimal solution even if it is not the best one. Additionally, the
DQC, we encounter similar optimization challenges as in the monolithic quantum mapping process can be separated into three processes:
case, but with the added complexity of distributing or cutting the
• Gate decomposition: Refers to the stage in which gates composing the
circuits. On the contrary, if the distribution technique performed is
circuit are transformed into a series of native gates implementable
embarrassingly parallel, the optimization phase is, naturally, equivalent
in the actual quantum processor. This is one of the aforementioned
to the monolithic one, excepting the case of multi-programming where
device’s constraints that have been taken into account.
optimizations are subtle and tend to be related with crosstalk and
fidelity [267,273]. • Quantum allocation: Refers to the process of physically assigning
Delving into circuit distribution, we have discussed in Section 4.1.1 specific logical qubits in a quantum processor. For a correct qubit
the circuit distribution methods and efforts made to partition the circuit allocation, in most cases, it is necessary to add additional SWAP gates
optimally before performing local mapping. In essence, optimization in to move the qubit information [304].
this case mirrors that of the monolithic case but with the additional • Quantum routing: Refers to the task of finding efficient paths for com-
consideration of the partitioning problem, which is intricately linked to munication between qubits in a quantum processor. This is important
qubit mapping. Indeed, the close relationship between qubit mapping when mapping gates of two logic qubits that are not interconnected
and circuit optimization is not surprising, even in the monolithic case. It to maximize efficiency [305,306]. For a thorough analysis of the
is logical because an efficient mapping of qubits directly impacts circuit qubit routing problem, one can check the review on the subject by
performance, much like how effective register management optimizes Barnes [307].
classical computing tasks. However, although we are only adding one
more constraint with the circuit distribution, it is of vital importance Regarding DQC, it is essential to distinguish between distribution
since the teleport and telegate costs are significantly higher than those methods that require partitioning and those that do not. In the former
of local 2-qubit gates. As previously discussed in Section 4.1.1, this case, where partitioning is necessary, the qubit mapping problem aligns
serves as justification for why circuit partitioning methods consistently with the classical problem. Still, it includes the additional challenge
aim to minimize the utilization of these remote protocols. Qiu and of optimizing circuit partitioning to minimize communication, as de-
Chen [300] realized an interesting analysis of this topic, where the tailed in Section 4.1.1, where we already mentioned how linked those
quantum cost figure of merit is employed. The quantum cost of a circuit methods are with this stage of compilation.
is calculated by summing the cost of each gate present in the circuit. Nevertheless, a few works that have not been mentioned in that
Any gate can be broken down into several basic gates, each with a section are of interest. The first one is the work of Mao et al. [308], who
unit cost, irrespective of their internal complexity. Using this definition named the problem as qubit allocation problem for distributed quantum
of cost, they showed the expensiveness of quantum teleportation and computing (QA-DQC), proved the NP-hardness of it and proposed two
dense coding. However, circling back to the main topic, while we have algorithms to deal with it: a heuristic local search algorithm and a
extensively covered and will further discuss partitioning in the qubit multistage hybrid simulated annealing (MHSA) algorithm. In the latter,
mapping section, we have deliberately chosen not to get deeply into they combine the local search algorithm and a simulated annealing
the intricate domain of monolithic quantum optimizations, as it exceeds meta-heuristic algorithm, along with extensive simulations to evaluate
the scope of this work. it. The second work was also carried out by Mao et al. [309] that

16
D. Barral et al. Computer Science Review 57 (2025) 100747

proposed a probability-aware qubit-to-processor mapping model, incor- This was further extended in a subsequent work by the same authors
porating communication overhead between processor pairs determined under the QuCloud+ framework [273], in which they tried to take into
through probabilistic analyses based on link entanglement generation consideration the crosstalk effect on real-world applications.
rates. Additionally, they introduced a multi-flow routing protocol to
Verification. The verification of quantum programs is a significant side
enhance overall entanglement rates. Subsequently, they employed a
aspect of quantum compiling. Unlike in the classical world, where
multistage hybrid simulated annealing algorithm, which is reminiscent
developers rely on debuggers to identify and fix errors, debugging
of the previous one, to minimize total communication overhead. As
quantum programs is inherently difficult due to the destructive na-
we have already mentioned, extensive simulations are conducted to
ture of measurement. Once a quantum state is measured, it collapses
demonstrate the effectiveness of these solutions across various system
irreversibly, making it impossible to observe the state at different
settings. The third work of interest in this line was the one developed by
time steps without altering it. Therefore, the verification of quantum
Nakai [310], which deeply developed the qubit allocation problem for programs becomes crucial for ensuring the correct functionality of a
DQC along with a formal definition of the problem as an optimization quantum circuit. It is essential to incorporate this verification step as a
problem similar to how we have defined the partitioning one. Finally, phase in the synthesis stage of compilation. This ensures that the circuit
the last work was developed by Chen et al. [311], where they focused is checked immediately before execution and after optimizations have
on the step following the circuit partitioning, i.e., the qubit routing been applied to confirm that those optimizations have not altered the
stage. Specifically, they focused on investigating the influence of the functionality of the quantum circuit. In the monolithic realm, several
quantum state transmission direction during the execution of global approaches have been made combining optimization and verification
gates on the number of transmissions and subsequent routing. It uti- in what is usually referred to as verified optimization [291,316,317].
lizes a heuristic algorithm, called Genetic Algorithm for Global Gate One way of verifying quantum programs is using quantum process
Direction Optimization (GAGDO), to ascertain the optimal transmission algebras, which are derivations of the classical process algebras. Process
direction for all global gates in the circuit, with the goal of minimizing algebras, also known as process calculi, are mathematically rigor-
the overall cost of the executable circuit generated in the distributed ous languages with well-defined semantics that allow the description
architecture model. and verification of properties of concurrent communicating systems,
Also, two works have been developed to characterize the inter- including, in this case, quantum systems.
core qubit traffic in which some benchmarks arise in order to analyze There are some examples of these types of formal methods. For
mapping performance [312,313]. They employed the OpenQL com- instance, Extended Quantum Process Algebra (eQPAlg) [318], which
piler [314], which is not a distributed compiler per se but allows the extends Quantum Process Algebra (QPAlg) [319]. More specifically,
embedding of a modified version of the Qmap mapper [315]. In par- QPAlg provides a homogeneous style for formal descriptions of con-
ticular, for this case, they extended it to the multi-core case employing current and distributed computations, encompassing both quantum and
the proposal by Baker et al. [204], i.e., the FGP-rOEE algorithm, already classical components. As authors claim, QPAlg introduces quantum
explained in Section 4.1.1. variables, operations on these variables – unitary operators and mea-
In the cases of embarrassingly parallel distribution that do not surement observables – as well as different forms of communication
require partitioning, the qubit mapping problem mirrors that of the involving the quantum realm. The operational semantics ensure that
monolithic case, with the added complexity of needing to perform these quantum objects, operations, and communications adhere to
mapping for each QPU. This complexity arises from the potential dif- the postulates of quantum mechanics. Regarding eQPAlg, it extends
ferences in architectures among the QPUs contained in the distributed the previous formal specification to accommodate the concept of for-
scheme. There is just one case in the embarrassingly parallel scenario mally specifying the quantum teleportation protocol, which has been
where qubit mapping differs from the monolithic case: the multi- shown in this work to be a key part of the quantum distribution
programming scenario. This paradigm of quantum execution, which model. The relationship between quantum process algebras and the
involves segmenting the QPU, imposes a series of constraints on the algebraic language defined in the aforementioned work by Ying and
qubit mapping problem. One of the first approaches was the already Feng [299] can be compared to that between classical process alge-
mentioned work by Das et al. [265]. Three techniques were developed bras and Boolean algebra. In broad terms, quantum process algebras
in this work: are well-suited for high-level formal specification of DQC, while the
language Ying and Feng paper is mainly intended to describe low-level
1. Fair and Reliable Partitioning (FRP) algorithms, developed to parti- circuit implementation.
tion qubit resources into multiple groups fairly while avoiding qubits Regarding the verification of distributed quantum programs, the
or links with excessively high error rates. work of Feng et al. [320] introduced a distributed programming lan-
2. Delayed Instruction Scheduling (DIS) policy, devised to mitigate guage designed for formalizing and verifying distributed quantum sys-
interference from measurement operations of one program on the tems. They presented a Hoare-style logic5 that is both sound and
gate operations of co-running programs. complete, aiding in the analysis and verification of quantum programs,
3. Adaptive Multi-Programming (AMP) design, proposed to monitor including quantum teleportation and CNOT gates. Talking specifically
reliability impact at runtime and revert the system to isolated ex- about distributed quantum protocols, Wang’s work [322] profoundly
ecution mode if the impact is high. delves into the verification of several distributed quantum protocols,
such as the BB84 protocol [323].
Different techniques were developed under the QuCloud framework
by Liu and Dou [267]. In this work, they also developed three ap- 4.2.4. Available compilers
proaches. First, they utilized community detection techniques to par- Not many full-stack tools or compilers are designed considering a
tition physical qubits among concurrent quantum programs, mitigat- distributed quantum scheme as a base. In fact, to the best of our knowl-
ing resource waste. They even proposed a new technology based on edge, there is no compiler for DQC available for use, just conceptual
these techniques called Community Detection Assistant Partitioning designs and prototypes. These conceptual quantum compilers can be
(CDAP). Second, they designed the X-SWAP scheme, which enables classified depending on which type of distribution they use from the
inter-program SWAPs and gives priority to SWAPs linked with critical ones described in Section 4.2, i.e., usual circuit distribution, circuit
gates to minimize SWAP overheads. Finally, they introduced a compi-
lation task scheduler that prioritizes the compilation and execution of
concurrent quantum programs based on estimated fidelity for optimal 5
Hoare logic is indeed a formal system equipped with a set of logical rules
performance. used for rigorous reasoning about the correctness of computer programs [321].

17
D. Barral et al. Computer Science Review 57 (2025) 100747

Table 2
Summary of available compilers for DQC, including their authors, reference, descriptions, main focus, and categorization by distribution type.
Category Tool/Compiler Authors Reference Available Main focus Description
Circuit Distributed Quantum Ferrari et al. [324] 3 Circuit depth minimization Designed to minimize circuit depth
distribution Compiler with specific partitioning using strategies based on data-qubit
strategies swapping and entanglement
swapping.
Compared against works like
Martínez and Heunen.
Circuit Modular Quantum Ferrari et al. [325] 3 Comprehensive optimization A modular framework considering
distribution Compilation Framework considering network, network and device constraints.
for DQC hardware, and specific Includes qubit assignment with
algorithms METIS, EPR pair minimization
algorithms, and optimized local
routing.
Circuit Cuomo’s compiler Cuomo et al. [326] 7 Optimization of distributed Models the compilation problem
distribution architectures for dynamic using Integer Linear Programming
networks and time-expanded network
representations.
Optimized for dynamic network
problems and quasi-parallelism.
Circuit cutting Qurzon (with CutQC) Chatterje [249] 3 t|𝑘𝑒𝑡⟩ for optimal qubit Employs CutQC for cutting circuits
et al. routing. Reconstructs original into optimal subcircuits without
circuit results quantum communications. Schedules
execution using a greedy algorithm.
Embarrassingly palloq Ohkura et al. [327] 3 High-fidelity layout synthesis Manages multi-programming with
parallel for multi- programming layout synthesis based on noise
scenarios adaptative layouts. Introduces a
crosstalk detection protocol and
integrates randomized benchmarking
for multi-circuit allocation.
Combining Quantum Divide and Tomesh et al. [328] 7 Hybrid variational approaches Combines circuit cutting and
techniques Conquer Algorithm combining cutting and distribution for hybrid variational
(QDCA) distribution applications. Uses graph partitioning
techniques like METIS and KL for
mapping large combinatorial
optimization problems to distributed
architectures.

cutting, and embarrassing parallelism. Table 2 provides a summary of also incorporates quasi-parallelism6 allows for more efficient circuit
the available compilers, detailing their authors, reference, descriptions, operation and broader solution exploration. The work is modular,
main focus, and categorization by distribution type, as discussed below. enabling adaptation to circuits with varying degrees of operation com-
mutativity and leveraging existing network flow literature. The study
Compilers for circuit distribution. Ferrari et al. [324] designed a dis-
aims to refine compiler efficiency and performance through an in-depth
tributed quantum compiler that focuses on the minimization of the
analysis of quantum circuits and focus on normal forms. Testing on
depth of the circuit and, for this matter, two different techniques
square and hexagonal lattice topologies, showed that square lattices
are tested: the data-qubit-swapping-based strategy and the entanglement-
offer superior performance, attributed to their favorable edges-to-nodes
swapping-based strategy. They compared the performance of the par-
ratio, indicating promising avenues for future quantum computing
titioning – and, hence, of the distribution – of these two strategies
advancements.
with the already analyzed work by Martinez and Heunen [189]. Also,
Ferrari et al. [325] designed a versatile modular quantum compila- Compilers for circuit cutting. As for now, the only quantum compiler con-
tion framework for DQC, which considers both network and device sidering the circuit-cutting strategy, as was explained in Section 4.1.2
constraints and characteristics. For qubit assignment, they employed is Qurzon [249]. For the first part of the compilation, an algorithm
METIS’s multilevel 𝑘-way partitioning. Moreover, for gate scheduling, responsible for cutting the circuit into optimal parts is employed, called
they implemented an algorithm to minimize the consumed EPR pairs CutQC [246]. After the circuit is cut into several pieces, a scheduling
and a local routing algorithm that scans the circuit and, for every gate algorithm is responsible for the execution of each of the pieces in the
that involves qubits not directly connected on their specific QPU, it available quantum devices. This problem is nothing more than a classic
computes the shortest sequence of necessary SWAP gates. The experi- problem of scheduling jobs, well known in the HPC environment. In
mental evaluation of a quantum compiler based on this framework was this case, a greedy algorithm is employed, at least in the theoretical
demonstrated, using circuits of interest such as VQE, QFT, and graph development of the compiler (since to obtain the results, they applied
state preparation, characterized by varying widths – ranging from 0 a so-called ‘‘naive’’ algorithm, which is not specified). For the optimal
up to 600 qubits. qubit routing, they reach out for the work of t|ket⟩ [329]. Then, a
Cuomo et al. [326] modeled the compilation problem using an distributed parallel execution is performed over the whole group of
Integer Linear Programming formulation inspired by the extensive subcircuits employing the different devices, and once the results are
theory on dynamic network problems. They defined the problem as
a generalization of the quickest multi-commodity flow, enabling opti-
mization using techniques from the literature, such as a time-expanded 6
The authors define quasi-parallelism as a relaxed version of parallelism
representation of the distributed architecture. This approach, which based on grouping logically sequenced gates within the same time step.

18
D. Barral et al. Computer Science Review 57 (2025) 100747

obtained, the CutQC work is again used to reconstruct the result of the
original circuit using every result obtained in each subcircuit.
Compilers for embarrassing parallelism. Despite the absence of compilers
specifically designed for embarrassingly parallel tasks in quantum com-
puting, the inherent parallelizable nature of these tasks – primarily the
distribution of shots across multiple QPUs – means that any quantum
compiler or framework could be easily modified to support this mode
of distribution. This adaptability is due to the fact that the distribution
of computational tasks among different processors is a well-established
practice in the field of HPC. Consequently, leveraging existing classi-
cal job distribution techniques allows for the straightforward parallel
execution of quantum computations on multiple QPUs, highlighting a
seamless integration of classical parallelism principles within quantum Fig. 12. Decision workflow for executing a quantum application in parallel.
computing frameworks.
Nevertheless, an appreciation of the multi-programming case has to
be made. Even though the already presented QuCloud and QuCloud+
Similar to classical HPC computing, the motivations for executing
[267,273] are considered mapping mechanisms, they possess a com-
an application in parallel may include insufficient resources on a single
pilation task scheduler and could be naturally extended to be able
QPU (e.g., a lack of qubits), stringent time constraints (where the time-
to perform as compilers with a multi-programming approach. This is
to-solution fails to meet the requirements of the intended use case),
precisely the scope of palloq, presented by Ohkura et al. [327], which
distributed input data across various storage locations (it is easier to
includes a layout synthesis for multiple quantum circuits and a job
move the computation close to the data than the reverse), concerns
scheduler to manage efficient and high fidelity quantum multi-pro-
over data security or confidentiality, among other considerations. The
gramming. This compiler takes multiple quantum circuits, written in
decision to parallelize can be made either by the user or delegated to
OpenQASM, and the device’s local gate error information as input.
an automatic scheduler.
Their layout synthesis employs a heuristic based on noise-adaptive
For instance, focusing solely on spatial constraints (i.e., the number
layout, where the device’s calibration data is analyzed to search for
of qubits) and the required shots to achieve an acceptable result, Fig. 12
improved allocation using a greedy approach. Additionally, they pro-
illustrates a decision workflow for selecting the appropriate execution
pose a software-based crosstalk detection protocol utilizing a novel
method for an application. If the circuit demands more qubits than are
combination of randomized benchmarking methods to characterize the
available on a single QPU and the program is parallelized, only the
hardware’s suitability for multi-programming.
required number of QPUs needs to be selected to execute it in parallel
Compilers combining types of distributions. The work from Tomesh et al. (e.g., circuit distribution, as outlined in Section 4.1). Conversely, if
[328] combines aspects of circuit distribution with the circuit-cutting the application is not parallelized, a preliminary step could involve
technique [328]. This work introduced an algorithm called Quantum assessing the feasibility of circuit cutting — dividing the problem into
Divide and Conquer Algorithm (QDCA), a hybrid variational approach multiple smaller circuits that can be executed independently without
aimed at mapping large combinatorial optimization problems onto quantum communication, which implies effectively transforming the
distributed quantum architectures. The QDCA specification contains task into an embarrassingly parallel process. Should circuit cutting
several key elements: the partition of the input combinatorial opti- prove insufficiently efficient, an autoparallelization mechanism may
mization problem into multiple subproblems, the construction of the reconfigure the circuit into a genuinely parallel program.
variational quantum circuit, and the execution of it on distributed quan- Conversely, if the basic circuit fits within the available QPU capac-
tum computers using quantum circuit cutting techniques. The partition ity, other options exist to use multiple QPUs in parallel to accelerate
of the input is where the classical techniques of graph partitioning execution. For instance, if the problem involves running several circuits
employed for circuit distribution take place, in this case, KL and METIS. that can be executed independently, such as during the optimization
Even though it is not circuit distribution per se, it employs the graph of a variational quantum circuit, these instances can be distributed
partitioning techniques used in this kind of distribution to perform across the available QPUs. Additionally, the same circuit instance can
circuit cutting, which narrows the boundaries between these two ap- be split among several QPUs, each handling only a fraction of the
proaches. This work presents quantum circuit cutting as a compilation required shots. However, if these circuits are interdependent, where
tool within a hybrid, variational application. With this approach, they the execution of one depends on the results of another, only shot
claimed to achieve approximate solutions to Maximum Independent Set parallelization is feasible.
(MIS) problems.7 The complexity of the decision workflow of Fig. 12 increases if
time constraints are included. In such cases, even if the program fits
5. Application layer within a single QPU, parallelization may become necessary to meet
the time requirements. However, selecting the number of QPUs must
This section explores proposed quantum applications that leverage be approached cautiously, as parallel execution inherently introduces
some of the methods previously outlined. Any quantum application overhead that must be accounted for.
executing at least one quantum circuit requiring multiple shots is In the remainder of this section, we will present some selected
inherently parallelizable, as the required shots can be distributed across examples of distributed quantum applications considering the division
available QPUs or the circuit can be partitioned using telegates or of Section 4.1 to show the possibilities of DQC. Specifically, we will dis-
teledata. However, it is important to note that such parallelization does cuss applications based on circuit distribution, those leveraging circuit
not necessarily guarantee enhanced performance; in fact, it could lead cutting, and others that can be formulated as embarrassingly parallel.
to a significant degradation. Other classifications of DQC applications also exist. For example, [330]
recently analyzed the applications of distributed quantum computing,
categorizing them into two main types: resource DQC, which addresses
7
The MIS problem is a classic NP-Complete combinatorial optimization scenarios where a single device lacks sufficient resources, and data
challenge defined on a graph 𝐺 = (𝑉 , 𝐸). Its objective is to identify the DQC, where data is distributed and QPUs can work collaboratively to
largest feasible independent set within 𝐺, where an independent set, denoted get the result. Within this framework, the authors review various ap-
as 𝑆 ⊂ 𝑉 , consists of nodes that are not adjacent to each other. plications and discuss the challenges of implementing them on current

19
D. Barral et al. Computer Science Review 57 (2025) 100747

hardware. Readers are encouraged to consult this article for further topologies. They found that, for these problems, the teledata out-
examples of applications leveraging DQC. performs the telegate method and that a linear architecture is the
best choice. In [336], Tan et al. described a parallel algorithm for
5.1. Circuit-distribution based applications Simon’s problem that still keeps the exponential scaling compared to
the classical algorithm.
As mentioned in the introduction of the paper, one of the first Recently, Li et al. [337] presented a family of distributed quantum
distributed algorithms was proposed by Grover [12]. In this work, he algorithms for the classical Deutsch-Jozsa problem. These algorithms
used a circuit distribution with quantum communications to estimate are based on a set of computers with remote communications. However,
the mean of 𝑁 numbers between −1 and 1 under ideal conditions. in the current description, the nature is still sequential, without a clear
Later, Gupta et al. [22] presented a distributed version of the Grover path to reduce the global depth and time. Finally, Shi et al. [284] made
search algorithm using quantum communications. Initially, the algo- a first proof of concept of using QMPI for the Quantum Phase Estima-
rithm was shown using only two QPUs, where an additional qubit was tion and Trotter time evolution, but without including real quantum
needed in each QPU to handle the quantum communications using an communications.
EPR pair. The complexity analysis showed that the classical Grover
requirements for operations are maintained in this distributed version 5.2. Circuit cutting and other hybrid applications
since the increase in the number of operations due to the distribution
scales with the number of qubits as in the original algorithm, but the As described in Section 4.1.2, algorithms based on circuit-cutting
number of classical communications per iteration is not increased. The only need classical communications to calculate the final solution.
paper did not show if the algorithm can scale to more than two QPUs. Automatic cutting of a circuit (in space or time) is feasible when the
One of the key quantum algorithms that present an exponential number of control operations to cut is limited. However, it is also
scaling is the Shor algorithm. The main drawback of this algorithm possible to use non-automatic clever designs to divide a single problem
is the high number of qubits that are needed for a correct execution. (usually executed using a single quantum circuit) in the execution
Due to this requirement, it is a perfect candidate to use the circuit of several independent quantum programs that later are combined
distribution technique. In [23], a first proposal to use several QPU was classically to find the right solution.
made. Firstly, they showed that the QFT could be executed in parallel, As already mentioned in the introduction, the paper from Yepez
substituting each controlled operation with a remote-controlled one, [24] was one of the first proposals to analyze this parallel computation
and that the modular exponentiation could be parallelized using a in a hybrid scheme. He considered the case of a system composed of
set of QPUs. Although a communication complexity of O((𝑙𝑜𝑔2 𝑁)2 ) is quantum nodes but exclusively connected by a classical network. He
needed, being 𝑁 the number of bits of the number to factorize, and the named this architecture type-II quantum architecture to differentiate
total number of qubits is increased, the size of each QPU is drastically it from the monolithic quantum processors (or type-I), which maintain
reduced. the global phase coherence. His proposal suggested that some problems
More recently, Gidney et al. [331] analyzed the hardware re- need only short spatial and time entanglement, as some kinds of
sources for factoring large numbers, using the Ekerå and Håstad al- molecules. So they are tractable in parallel quantum computers, unlike
gorithm [332] instead of the Shor one. Applying several optimizations other algorithms that need long and spatially large entanglement. For
and considering the current methods for making logical qubits, they solving those problems, there are three assumptions: first, that the wave
asserted that a number of 2048 bits can be factorized in 8 h with 20 function is separable, i.e., can be expressed as a tensor product of
million noisy qubits (if the operations work in the range of nanosec- subwave functions, each of them residing in one QPU; second, that we
onds). However, due to the capabilities of the implemented additions can apply a projection operator simultaneously on each qubit of each
needed to factorize the number, the qubits can be reduced to 11 million QPU; and, third, that this projection can be applied after each time
for each QPU when 2 are used and to 4 million for 8 QPUs. They require step. Yepez proposes a quantum computer composed of many small
a quantum network with a low (but efficient) bandwidth of 150 qb/s. QPUs arranged in a regular periodic lattice, where local operations are
Later, Xiao et al. [333] presented a parallel algorithm that reduces the applied to the local qubits simultaneously across the lattice. He applies
number of needed qubits, dividing the algorithm between several QPUs, this proposal to solve problems with lattice gases. For small QPUs, the
each calculating one subset of the bits. Although it uses several QPUs, problems could be tractable using modern Tensor Networks techniques.
it is sequential because to guarantee that the correct state is used on In [338,339], Zhou et al. presented distributed quantum algorithms
each step, it is teletransported between them at the end of each step. for the Bernstein-Vazirani classical problem and the Grover search,
More well-known quantum algorithms have been parallelized. For respectively. They divide the binary functions used in the algorithms
example, Neumann et al. [334] studied the Quantum Phase Estimation into a set of subfunctions that can be executed in parallel, getting
algorithm using a remote-controlled operation. They compared two the final result by composing the different binary parts. In the case
possible approaches. The first one is called standard (or automatic), of Grover’s search, the algorithm only works when a single solution
where each controlled operation in the standard QFT is replaced by exists, while the extension is still open to multiple solutions. Similarly,
a remote-controlled operation. This case needs 𝑛2 entangled pairs to Avron et al. [340] studied Deutsch-Jozsa’s, Simon’s, and Grover’s on a
execute. The second approach uses the iterative nature of the QFT, distributed environment, finding that, for these algorithms, there are
aggregating all controlled operations by a single qubit in a unique still advantages when comparing with the classical solutions, being the
transport operation. In this case, the number of transport operations advantage reduced when compared with the fault-tolerant versions. But
is reduced to 𝑛. They used a simulator for the experiments, introducing since these distributed algorithms require shallow circuits, they may be
different noise levels to create entangled pairs. The results obtained a short-term solution in today’s NISQ era.
are similar for both approaches, given the last systematically better Several parallel versions of VQAs also use circuit-cutting techniques.
results. This experiment showed that automatic partitioning of the For example, [224] used a circuit-cutting based VQE to calculate the
problems must take care of possible optimizations and multiple usage ground state of BeH2 . Eddins et al. [221] presented another kind of
of a single pair. One important point is that they studied only the methodology. They use the Schmidt decomposition to divide a chemical
effect of imperfect entangling in the needed pairs without considering problem of 2𝑁 qubits in several circuits that need only 𝑁 qubits, ap-
other errors such as the measurement, controlled operations between plying VQE to those and joining the results to calculate the final value
the pairs, and the QPU qubits, etc. of the observable. Fujii et al. [341] proposed another method to divide
Also, Van Meter et al. [335] studied some of the possible arithmetic the problem into smaller cases that are combined hierarchically to find
operations using teledata and telegate methods in different distributed the final solution. The technique can be applied when the problem has

20
D. Barral et al. Computer Science Review 57 (2025) 100747

some structure that aggregates the entanglement in clusters that can be • Quantum Networks: To achieve interconnected, datacenter-scale
linked later at a higher level. They applied the technique to a kagome QPUs, quantum networks must enable entanglement distribution be-
lattice, using several layers of aggregation. This technique could also tween any two nodes in the network. Current scalable proposals sug-
be used in a hybrid scheme, where part of the calculation is done by gest using quantum networking devices such as repeaters, switches,
QPUs at the first steps, and later, the system is solved by a classical and routers. These devices support the pre-establishment of entan-
computer using tensor networks. gled qubits through transduction to flying qubits and successive
The usage of these divide-and-conquer techniques can also be ap- entanglement distribution to end nodes, where computation occurs.
plied to combinatorial optimization, where a larger problem can be
Open challenges: Such devices must include registers of qubits and
solved using several computers [328,342], and to Quantum Machine
implement a limited quantum operation instruction set to execute en-
Learning (QML). Marshall et al. [343] examine it for the case of
tanglement distillation, swapping, and teleportation protocols. These
classification. They found that automatic circuit cutting could avoid
advancements are essential to unlock true deterministic DQC archi-
executing all the subcircuits because some of them do not contribute
tectures. Alternative approaches based on teledata operation with
significantly to the final result and proposed a small change in the
single flying qubits instead of EPR pairs could simplify network
process that permits the achievement of results close to the classical
architectures. However, further research is required to match the fi-
Neural Networks.
delity and efficiency of current entanglement-based protocols. On the
other hand, from a practical and market-oriented perspective, current
5.3. Embarrassingly parallel applications
quantum networking solutions are costly and lack the required per-
The cutting techniques presented in the previous section convert formance, fidelity, and robustness. Higher-level aspects remain in the
a complex problem into an example of an embarrassingly parallel early stages of research, including developing networking protocols,
application, where each smaller circuit can be executed in parallel, scalable connectivity architectures, and robust systems. Auxiliary pro-
later combining the results classically. Other examples of these kinds of tocols for synchronization, resource management for entanglement
applications are [344,345], which studied the use of partial diffusion distribution, network service definition, error correction, and qubit
operator [346] for Grover’s search algorithm. The use of this technique encoding must still be developed to achieve fault-tolerant, highly
does not reduce the number of required qubits but presents some available, and performant quantum networks suitable for distributed
advantages because each circuit is smaller in depth (and, consequently, quantum computing.
needs less time to execute in parallel), and the angles of rotations are • Circuit Cutting: In the current noisy and limited QPUs scenario,
bigger, reducing the errors in current quantum devices. circuit cutting can be a useful tool for solving large problems with
Other quantum algorithms, such as the Phase Estimation for a small quantum computers by distributing parts of the circuit between
single phase, can be executed using this formalism [347] because it is them without requiring a fully realized quantum network.
possible to split the algorithm into several smaller circuits and combine Open challenges: The cost associated with this technique scales expo-
the results classically at the end. Other classical quantum algorithms, nentially with the amount of cut entanglement between the parts,
such as the Amplitude Estimation, require large resources that can be and, for general quantum circuits, entanglement may have a very
approximated by distributing several smaller tasks and post-processing complex structure that is unknown beforehand. Some improvements
classically their results [348]. have been proposed, which could avoid the execution of a large
In order to get the maximum profit from the available distributed fraction of the subcircuits, thereby reducing the computing require-
infrastructure or, in the short term, to permit the calculation of VQAs, ment. Nonetheless, there are criticisms about the overall utility of
a combination of the aforementioned techniques can be applied. For these techniques. Moreover, dividing circuits and executing them
instance, DiAdamo et al. [349] proposed placing some circuits needed on different QPUs requires a better understanding of the effect of
for calculating the expectation value on available QPUs, using the different noise profiles on each QPU. Additionally, when different
remaining free qubits to create a distributed version of the Ansatz. Al- architectures are employed, the execution times must be carefully
ternatively, the Ansatz could be split using the circuit cutting technique. managed.
• Compilers: Using agnostic compilers to find the best partitions for
6. Final remarks and open challenges a general algorithm is similar to auto-parallelism in classical com-
puting, which scales poorly. Designing problems that are easier to
Distributed quantum computing emerges as a clear pathway to partition, such as well-designed ansatzes for variational quantum al-
enhance the computational capabilities of current quantum systems. In
gorithms or problems tailored for modular architectures, may be more
this work, we have presented a comprehensive survey of this field’s
effective. In addition to automatic circuit-breaking tools, experienced
current state of the art. Using a four-layered model – physical, network,
programmers can develop methods for dividing and parallelizing
development, and application –, we have guided readers to explore its
algorithms. Tools like QMPI or frameworks for distributing programs
foundational principles, achievements, challenges, and promising direc-
are also necessary.
tions for further research. Next, we conclude this work by summarizing
some of the most important open challenges in the DQC field: Open challenges: Research is needed to improve agnostic compil-
ers, develop more efficient partitioning methods, and create tools
• Quantum Teleportation: It is the most fundamental mechanism that enable programmers to parallelize quantum computations across
required at the physical layer for distributed algorithms in DQC different quantum processors efficiently.
applications. Two types of teleportation protocols are essential: gate • Applications: Embarrassingly parallel applications or those based on
teleportation (telegate) and qubit state teleportation (teledata). Tel- circuit knitting are the most widely used solutions in the current NISQ
egate enables the remote execution of quantum gates on entangled era.
qubits, allowing quantum information to be manipulated without
direct physical interaction. Teledata allows an unknown quantum Open challenges: Further research is needed to develop high-level
state processed at one network node to be transmitted to a remote parallel programming models for distributed quantum computing that
location. efficiently use future quantum networks.

Open challenges: Enhancing the fidelity of these protocols is an active It can be concluded from this work that distributed quantum com-
area of research, as high fidelity is critical for ensuring quantum- puting offers a promising way to overcome the limitations of cur-
computational accuracy in future distributed quantum computers. rent quantum systems by connecting and scaling quantum processors.

21
D. Barral et al. Computer Science Review 57 (2025) 100747

While significant challenges remain – such as improving teleporta- [9] A.J. McCaskey, T. Nguyen, A. Santana, D. Claudino, T. Kharazi, H. Finkel,
tion fidelity, developing scalable networks and optimizing compilers – Extending C++ for heterogeneous quantum-classical computing, ACM Trans.
Quantum Comput. 2 (2) (2021) 1–36, http://dx.doi.org/10.1145/3462670.
advances in these areas will facilitate the path towards robust and fault-
[10] J. Gambetta, Quantum-centric supercomputing: The next wave of computing,
tolerant quantum computing, unlocking unprecedented computational 2022, URL https://www.ibm.com/quantum/blog/next-wave-quantum-centric-
capabilities. supercomputing. (Accessed 21 January 2025).
[11] J. Gambetta, The hardware and software for the era of quantum utility is here,
Declaration of Generative AI and AI-assisted technologies in the 2023, URL https://research.ibm.com/blog/quantum-roadmap-2033. (Accessed
writing process 21 January 2025).
[12] L.K. Grover, Quantum telecomputation, 1997, http://dx.doi.org/10.48550/
arXiv.quant-ph/9704012, arXiv.
During the preparation of this work the authors used ChatGPT
[13] R. Cleve, H. Buhrman, Substituting quantum entanglement for communication,
in order to improve language and readability. After using this tool, Phys. Rev. A 56 (2) (1997) 1201–1204, http://dx.doi.org/10.1103/PhysRevA.
the authors reviewed and edited the content as needed and take full 56.1201.
responsibility for the content of the publication. [14] J.I. Cirac, A.K. Ekert, S.F. Huelga, C. Macchiavello, Distributed quantum
computation over noisy channels, Phys. Rev. A 59 (1999) 4249–4254, http:
//dx.doi.org/10.1103/PhysRevA.59.4249.
Declaration of competing interest [15] J. Eisert, K. Jacobs, P. Papadopoulos, M.B. Plenio, Optimal local implementation
of nonlocal quantum gates, Phys. Rev. A 62 (5) (2000) http://dx.doi.org/10.
The authors declare the following financial interests/personal rela- 1103/PhysRevA.62.052317.
tionships which may be considered as potential competing interests: [16] D. Collins, N. Linden, S. Popescu, Nonlocal content of quantum operations,
Phys. Rev. A - At. Mol. Opt. Phys. 64 (2001) 7, http://dx.doi.org/10.1103/
Tomas F. Pena reports financial support and article publishing charges
PhysRevA.64.032302.
were provided by European Union. Tomas F. Pena reports financial [17] D.P. DiVincenzo, The physical implementation of quantum computa-
support was provided by Goverment of Spain MINECO. Tomas F. tion, Fortschr. Phys. 48 (2000) 771–783, http://dx.doi.org/10.1002/1521-
Pena reports financial support was provided by Government of Galicia. 3978(200009)48:9/11<771::AID-PROP771>3.0.CO;2-E.
If there are other authors, they declare that they have no known [18] Y.L. Lim, A. Beige, L.C. Kwek, Repeat-until-success linear optics distributed
quantum computing, Phys. Rev. Lett. 95 (2005) http://dx.doi.org/10.1103/
competing financial interests or personal relationships that could have
PhysRevLett.95.030505.
appeared to influence the work reported in this paper. [19] A. Serafini, S. Mancini, S. Bose, Distributed quantum computation via optical
fibres, Phys. Rev. Lett. 96 (2006) http://dx.doi.org/10.1103/PhysRevLett.96.
Acknowledgments 010503.
[20] L. Jiang, J.M. Taylor, A.S. Sørensen, M.D. Lukin, Distributed quantum compu-
tation based on small quantum registers, Phys. Rev. A - At. Mol. Opt. Phys. 76
This work was supported by MICINN through the European Union
(6) (2007) http://dx.doi.org/10.1103/PhysRevA.76.062323.
NextGenerationEU recovery plan (PRTR-C17.I1), the Galician Regional [21] D.K.L. Oi, S.J. Devitt, L.C.L. Hollenberg, Scalable error correction in dis-
Government through ‘‘Planes Complementarios de I+D+I con las Co- tributed ion trap computers, Phys. Rev. A 74 (2006) http://dx.doi.org/10.1103/
munidades Autónomas’’ in Quantum Communication, MINECO (grants physreva.74.052313.
PID2019-104834GB-I00, PID2022-141623NB-I00 and PID2022- [22] M. Gupta, A. Pathak, A scheme for distributed quantum search through simul-
taneous state transfer mechanism, Ann. Phys., Lpz. 16 (12) (2007) 791–797,
137061OB- C22), Consellería de Cultura, Educación e Ordenación
http://dx.doi.org/10.1002/andp.200710265.
Universitaria (accreditations ED431C 2022/16 and ED431G-2019/04), [23] A. Yimsiriwattana, S.J. Lomonaco Jr., Distributed quantum computing: A
and the European Regional Development Fund (ERDF). distributed shor algorithm, Proc. SPIE 5436 (2004) 360–372, http://dx.doi.org/
10.1117/12.546504.
Data availability [24] J. Yepez, Type-II quantum computers, Internat. J. Modern Phys. C 12 (9) (2001)
1273–1284, http://dx.doi.org/10.1142/S0129183101002668.
[25] R. Jozsa, N. Linden, On the role of entanglement in quantum-computational
No data was used for the research described in the article. speed-up, Proc. R. Soc. A: Math. Phys. Eng. Sci. 459 (2003) 2011–2032,
http://dx.doi.org/10.1098/rspa.2002.1097.
[26] M. Caleffi, M. Amoretti, D. Ferrari, J. Illiano, A. Manzalini, A.S. Cacciapuoti,
References Distributed quantum computing: a survey, Comput. Netw. 254 (2024) 110672,
http://dx.doi.org/10.1016/j.comnet.2024.110672.
[1] R. Van Meter, W.J. Munro, K. Nemoto, Architecture of a quantum multicom- [27] S. Rodrigo, S. Abadal, E. Alarcon, M. Bandic, H.V. Someren, C.G. Almudever, On
puter implementing shor’s algorithm, in: Theory of Quantum Computation, double full-stack communication-enabled architectures for multicore quantum
Communication, and Cryptography, Springer, 2008, pp. 105–114, http://dx. computers, IEEE Micro 41 (5) (2021) 48–56, http://dx.doi.org/10.1109/MM.
doi.org/10.1007/978-3-540-89304-2_10. 2021.3092706.
[2] QuEra, Benefits of modular quantum computing for business, 2023, URL [28] D. Cuomo, M. Caleffi, A.S. Cacciapuoti, Towards a distributed quantum com-
https://www.quera.com/blog-posts/benefits-of-modular-quantum-computing- puting ecosystem, IET Quantum Commun. 1 (1) (2020) 3–8, http://dx.doi.org/
for-business. (Accessed 21 January 2025). 10.1049/iet-qtc.2020.0002.
[3] Quantum news, Distributed quantum computing: Scaling quantum [29] W.K. Wootters, W.H. Zurek, A single quantum cannot be cloned, Nature 299
power with multiple processors and noise simulation, 2024, URL (5886) (1982) 802–803, http://dx.doi.org/10.1038/299802a0.
https://quantumzeitgeist.com/distributed-quantum-computing-scaling-quantum- [30] R. Horodecki, P. Horodecki, M. Horodecki, K. Horodecki, Quantum entan-
power-with-multiple-processors-and-noise-simulation. (Accessed 21 January glement, Rev. Modern Phys. 81 (2009) 865–942, http://dx.doi.org/10.1103/
2025). RevModPhys.81.865.
[4] IonQ, IonQ achieves critical first step towards developing future quantum [31] C.C. Gerry, P.L. Knight, Introductory Quantum Optics, Cambridge Universiy
networks, 2024, URL https://ionq.com/news/ionq-achieves-critical-first-step- Press, 2005, http://dx.doi.org/10.1017/CBO9780511791239.
towards-developing-future-quantum-networks. (Accessed 21 January 2025). [32] P. Andres-Martinez, Towards Distributed Quantum Algorithms (Master’s thesis),
[5] N. Saurabh, S. Jha, A. Luckow, A conceptual architecture for a Quantum- School of Informatics, University of Edinburgh, 2018, URL https://project-
HPC middleware, in: Int. Conf. on Quantum Software, QSW, IEEE, 2023, pp. archive.inf.ed.ac.uk/msc/20183076/msc_proj.pdf.
116–127, http://dx.doi.org/10.1109/QSW59989.2023.00023. [33] C.H. Bennett, G. Brassard, C. Crépeau, R. Jozsa, A. Peres, W.K. Wootters,
[6] K. Wintersperger, H. Safi, W. Mauerer, QPU-system co-design for quantum Teleporting an unknown quantum state via dual classical and Einstein-Podolsky-
HPC accelerators, in: Architecture of Computing Systems, ARCS 2022, in: Rosen channels, Phys. Rev. Lett. 70 (13) (1993) 1895–1899, http://dx.doi.org/
Lecture Notes in Computer Science, vol. 13642, Springer, 2022, pp. 100–114, 10.1103/PhysRevLett.70.1895.
http://dx.doi.org/10.1007/978-3-031-21867-5_7. [34] A. Acín, I. Bloch, H. Buhrman, T. Calarco, C. Eichler, J. Eisert, D. Esteve,
[7] J. Vázquez-Pérez, C. Piñeiro, J.C. Pichel, T.F. Pena, A. Gómez, QPU integration N. Gisin, S.J. Glaser, F. Jelezko, S. Kuhr, M. Lewenstein, M.F. Riedel, P.O.
in OpenCL for heterogeneous programming, J. Supercomput. (2024) 1–22, Schmidt, R. Thew, A. Wallraff, I. Walmsley, F.K. Wilhelm, The quantum
http://dx.doi.org/10.1007/s11227-023-05879-9. technologies roadmap: A European community view, New J. Phys. 20 (2018)
[8] A.J. McCaskey, D.I. Lyakh, E.F. Dumitrescu, S.S. Powers, T.S. Humble, http://dx.doi.org/10.1088/1367-2630/aad1ea.
XACC: A system-level software infrastructure for heterogeneous quantum- [35] A. Karlsson, M. Bourennane, Quantum teleportation using three-particle en-
classical computing, 2019, http://dx.doi.org/10.48550/arXiv.1911.02452, arxiv tanglement, Phys. Rev. A 58 (1998) 4394–4400, http://dx.doi.org/10.1103/
preprint. PhysRevA.58.4394.

22
D. Barral et al. Computer Science Review 57 (2025) 100747

[36] M. Żukowski, A. Zeilinger, M.A. Horne, A.K. Ekert, ‘‘Event-ready-detectors’’ A. Zalcman, Y. Zhang, N. Zhu, N. Zobrist, H. Neven, R. Babbush, D. Bacon, S.
bell experiment via entanglement swapping, Phys. Rev. Lett. 71 (26) (1993) Boixo, J. Hilton, E. Lucero, A. Megrant, J. Kelly, Y. Chen, V. Smelyanskiy,
4287–4290, http://dx.doi.org/10.1103/PhysRevLett.71.4287. X. Mi, V. Khemani, P. Roushan, G.Q. AI, Collaborators, Measurement-induced
[37] H.-J. Briegel, W. Dür, J.I. Cirac, P. Zoller, Quantum repeaters: The role of entanglement and teleportation on a noisy quantum processor, Nature 622
imperfect local operations in quantum communication, Phys. Rev. Lett. 81 (7983) (2023) 481–486, http://dx.doi.org/10.1038/s41586-023-06505-7.
(1998) 5932–5935, http://dx.doi.org/10.1103/PhysRevLett.81.5932. [53] A. Yimsiriwattana, S.J. Lomonaco Jr., Generalized GHZ states and dis-
[38] J.-G. Ren, P. Xu, H.-L. Yong, L. Zhang, S.-K. Liao, J. Yin, W.-Y. Liu, W.-Q. Cai, tributed quantum computing, 2004, http://dx.doi.org/10.48550/arXiv.quant-
M. Yang, L. Li, K.-X. Yang, X. Han, Y.-Q. Yao, J. Li, H.-Y. Wu, S. Wan, L. Liu, ph/0402148, arXiv.
D.-Q. Liu, Y.-W. Kuang, Z.-P. He, P. Shang, C. Guo, R.-H. Zheng, K. Tian, Z.-C. [54] C.H. Bennett, D.P. DiVincenzo, P.W. Shor, J.A. Smolin, B.M. Terhal, W.K.
Zhu, N.-L. Liu, C.-Y. Lu, R. Shu, Y.-A. Chen, C.-Z. Peng, J.-Y. Wang, J.-W. Pan, Wootters, Remote state preparation, Phys. Rev. Lett. 87 (2001) http://dx.doi.
Ground-to-satellite quantum teleportation, Nature 549 (7670) (2017) 70–73, org/10.1103/PhysRevLett.87.077902.
http://dx.doi.org/10.1038/nature23675. [55] H. Weinfurter, Experimental bell-state analysis, Europhys. Lett. 25 (1994)
[39] D. Gottesman, I.L. Chuang, Demonstrating the viability of universal quantum 559–564, http://dx.doi.org/10.1209/0295-5075/25/8/001.
computation using teleportation and single-qubit operations, Nature 402 (6760) [56] S. Massar, S. Popescu, Optimal extraction of information from finite quantum
(1999) 390–393, http://dx.doi.org/10.1038/46503. ensembles, Phys. Rev. Lett. 74 (1995) 1259–1263, http://dx.doi.org/10.1103/
[40] R. Raussendorf, H.J. Briegel, A one-way quantum computer, Phys. Rev. Lett. PhysRevLett.74.1259.
86 (2001) 5188–5191, http://dx.doi.org/10.1103/PhysRevLett.86.5188. [57] S. Pirandola, J. Eisert, C. Weedbrook, A. Furusawa, S.L. Braunstein, Advances
[41] K.S. Chou, J.Z. Blumoff, C.S. Wang, P.C. Reinhold, C.J. Axline, Y.Y. Gao, L. in quantum teleportation, Nat. Photonics 9 (2015) 641–652, http://dx.doi.org/
Frunzio, M.H. Devoret, L. Jiang, R.J. Schoelkopf, Deterministic teleportation of 10.1038/nphoton.2015.154.
a quantum gate between two logical qubits, Nature 561 (7723) (2018) 368–373, [58] X.M. Hu, Y. Guo, B.H. Liu, C.F. Li, G.C. Guo, Progress in quantum teleportation,
http://dx.doi.org/10.1038/s41586-018-0470-y. Nat. Rev. Phys. 5 (2023) 339–353, http://dx.doi.org/10.1038/s42254-023-
[42] D. Bouwmeester, J.-W. Pan, K. Mattle, M. Eibl, H. Weinfurter, A. Zeilinger, 00588-x.
Experimental quantum teleportation, Nature 390 (6660) (1997) 575–579, http: [59] Q.C. Sun, Y.L. Mao, S.J. Chen, W. Zhang, Y.F. Jiang, Y.B. Zhang, W.J. Zhang,
//dx.doi.org/10.1038/37539. S. Miki, T. Yamashita, H. Terai, X. Jiang, T.Y. Chen, L.X. You, X.F. Chen, Z.
[43] A. Furusawa, J.L. Sørensen, S.L. Braunstein, C.A. Fuchs, H.J. Kimble, E.S. Polzik, Wang, J.Y. Fan, Q. Zhang, J.W. Pan, Quantum teleportation with independent
Unconditional quantum teleportation, Science 282 (5389) (1998) 706–709, sources and prior entanglement distribution over a network, Nat. Photonics 10
http://dx.doi.org/10.1126/science.282.5389.706. (2016) 671–675, http://dx.doi.org/10.1038/nphoton.2016.179.
[44] J.F. Sherson, H. Krauter, R.K. Olsson, B. Julsgaard, K. Hammerer, I. Cirac, E.S.
[60] R. Valivarthi, M.G. Puigibert, Q. Zhou, G.H. Aguilar, V.B. Verma, F. Marsili,
Polzik, Quantum teleportation between light and matter, Nature 443 (2006)
M.D. Shaw, S.W. Nam, D. Oblak, W. Tittel, Quantum teleportation across a
557–560, http://dx.doi.org/10.1038/nature05136.
metropolitan fibre network, Nat. Photonics 10 (2016) 676–680, http://dx.doi.
[45] M. Nielsen, E. Knill, R. Laflamme, Complete quantum teleportation using org/10.1038/nphoton.2016.180.
nuclear magnetic resonance, Nature 396 (1998) 52–55, http://dx.doi.org/10.
[61] C.M. Knaut, A. Suleymanzade, Y.-C. Wei, D.R. Assumpcao, P.-J. Stas, Y.Q. Huan,
1038/23891.
B. Machielse, E.N. Knall, M. Sutula, G. Baranes, N. Sinclair, C. De-Eknamkul,
[46] M. Riebe, H. Häffner, C.F. Roos, W. Hänsel, J. Benheim, G.P. Lancaster, T.W.
D.S. Levonian, M.K. Bhaskar, H. Park, M. Lončar, M.D. Lukin, Entanglement of
Körber, C. Becher, F. Schmidt-Kaler, D.F. James, R. Blatt, Deterministic quantum
nanophotonic quantum memory nodes in a telecommunication network, 2023,
teleportation with atoms, Nature 429 (2004) 734–737, http://dx.doi.org/10.
http://dx.doi.org/10.48550/arXiv.2310.01316, arXiv eprint.
1038/nature02570.
[62] V. Krutyanskiy, M. Galli, V. Krcmarsky, S. Baier, D.A. Fioretto, Y. Pu, A.
[47] M.D. Barrett, J. Chiaverini, T. Schaetz, J. Britton, W.M. Itano, J.D. Jost, E.
Mazloom, P. Sekatski, M. Canteri, M. Teller, J. Schupp, J. Bate, M. Meraner,
Knill, C. Langer, D. Leibfried, R. Ozeri, D.J. Wineland, Deterministic quantum
N. Sangouard, B.P. Lanyon, T.E. Northup, Entanglement of trapped-ion qubits
teleportation of atomic qubits, Nature 429 (2004) 737–739, http://dx.doi.org/
separated by 230 meters, Phys. Rev. Lett. 130 (2023) http://dx.doi.org/10.
10.1038/nature02608.
1103/PhysRevLett.130.050803.
[48] L. Steffen, Y. Salathe, M. Oppliger, P. Kurpiers, M. Baur, C. Lang, C. Eichler, G.
[63] J.-L. Liu, X.-Y. Luo, Y. Yu, C.-Y. Wang, B. Wang, Y. Hu, J. Li, M.-Y. Zheng, B.
Puebla-Hellmann, A. Fedorov, A. Wallraff, Deterministic quantum teleportation
Yao, Z. Yan, D. Teng, J.-W. Jiang, X.-B. Liu, X.-P. Xie, J. Zhang, Q.-H. Mao, X.
with feed-forward in a solid state system, Nature 500 (2013) 319–322, http:
Jiang, Q. Zhang, X.-H. Bao, J.-W. Pan, A multinode quantum network over a
//dx.doi.org/10.1038/nature12422.
metropolitan area, 2023, http://dx.doi.org/10.48550/arXiv.2309.00221, arXiv
[49] X.L. Wang, X.D. Cai, Z.E. Su, M.C. Chen, D. Wu, L. Li, N.L. Liu, C.Y. Lu, J.W.
eprint.
Pan, Quantum teleportation of multiple degrees of freedom of a single photon,
[64] S. Daiss, S. Langenfeld, S. Welte, E. Distante, P. Thomas, L. Hartung, O.
Nature 518 (2015) 516–519, http://dx.doi.org/10.1038/nature14246.
Morin, G. Rempe, A quantum-logic gate between distant quantum-network mod-
[50] X.M. Hu, C. Zhang, B.H. Liu, Y. Cai, X.J. Ye, Y. Guo, W.B. Xing, C.X. Huang,
ules, Science 371 (6529) (2021) 614–617, http://dx.doi.org/10.1126/science.
Y.F. Huang, C.F. Li, G.C. Guo, Experimental high-dimensional quantum telepor-
abe3150.
tation, Phys. Rev. Lett. 125 (2020) http://dx.doi.org/10.1103/PhysRevLett.125.
[65] S. Langenfeld, S. Welte, L. Hartung, S. Daiss, P. Thomas, O. Morin, E. Distante,
230501.
G. Rempe, Quantum teleportation between remote qubit memories with only
[51] D. Llewellyn, Y. Ding, I.I. Faruque, S. Paesani, D. Bacco, R. Santagati, Y.J.
a single photon as a resource, Phys. Rev. Lett. 126 (2021) 130502, http:
Qian, Y. Li, Y.F. Xiao, M. Huber, M. Malik, G.F. Sinclair, X. Zhou, K. Rottwitt,
//dx.doi.org/10.1103/PhysRevLett.126.130502.
J.L. O’Brien, J.G. Rarity, Q. Gong, L.K. Oxenlowe, J. Wang, M.G. Thompson,
Chip-to-chip quantum teleportation and multi-photon entanglement in silicon, [66] Y. Wan, D. Kienzler, S. Erickson, K. Mayer, T. Tan, J. Wu, H. Vasconcelos,
Nat. Phys. 16 (2020) 148–153, http://dx.doi.org/10.1038/s41567-019-0727-x. S. Glancy, E. Knill, D. Wineland, A. Wilson, D. Leibfried, Quantum gate
[52] J.C. Hoke, M. Ippoliti, E. Rosenberg, D. Abanin, R. Acharya, T.I. Andersen, M. teleportation between separated qubits in a trapped-ion processor, Science 364
Ansmann, F. Arute, K. Arya, A. Asfaw, J. Atalaya, J.C. Bardin, A. Bengtsson, (2019) 875–878, http://dx.doi.org/10.1126/science.aaw9415.
G. Bortoli, A. Bourassa, J. Bovaird, L. Brill, M. Broughton, B.B. Buckley, D.A. [67] D. Lago-Rivera, S. Grandi, J. Rakonjac, A. Seri, H. de Riedmatten, Telecom-
Buell, T. Burger, B. Burkett, N. Bushnell, Z. Chen, B. Chiaro, D. Chik, J. heralded entanglement between multimode solid-state quantum memories,
Cogan, R. Collins, P. Conner, W. Courtney, A.L. Crook, B. Curtin, A.G. Dau, Nature 594 (2021) 37–40, http://dx.doi.org/10.1038/s41586-021-03481-8.
D.M. Debroy, A. Del Toro Barba, S. Demura, A. Di Paolo, I.K. Drozdov, A. [68] C.H. Bennett, G. Brassard, S. Popescu, B. Schumacher, J.A. Smolin, W.K.
Dunsworth, D. Eppens, C. Erickson, E. Farhi, R. Fatemi, V.S. Ferreira, L.F. Wootters, Purification of noisy entanglement and faithful teleportation via
Burgos, E. Forati, A.G. Fowler, B. Foxen, W. Giang, C. Gidney, D. Gilboa, noisy channels, Phys. Rev. Lett. 76 (1996) 722–725, http://dx.doi.org/10.1103/
M. Giustina, R. Gosula, J.A. Gross, S. Habegger, M.C. Hamilton, M. Hansen, PhysRevLett.76.722.
M.P. Harrigan, S.D. Harrington, P. Heu, M.R. Hoffmann, S. Hong, T. Huang, [69] D.E. Browne, T. Rudolph, Resource-efficient linear optical quantum compu-
A. Huff, W.J. Huggins, S.V. Isakov, J. Iveland, E. Jeffrey, Z. Jiang, C. Jones, tation, Phys. Rev. Lett. 95 (2005) http://dx.doi.org/10.1103/PhysRevLett.95.
P. Juhas, D. Kafri, K. Kechedzhi, T. Khattar, M. Khezri, M. Kieferová, S. 010501.
Kim, A. Kitaev, P.V. Klimov, A.R. Klots, A.N. Korotkov, F. Kostritsa, J.M. [70] S. Bartolucci, P. Birchall, H. Bombín, H. Cable, C. Dawson, M. Gimeno-Segovia,
Kreikebaum, D. Landhuis, P. Laptev, K.-M. Lau, L. Laws, J. Lee, K.W. Lee, Y.D. E. Johnston, K. Kieling, N. Nickerson, M. Pant, F. Pastawski, T. Rudolph,
Lensky, B.J. Lester, A.T. Lill, W. Liu, A. Locharla, O. Martin, J.R. McClean, C. Sparrow, Fusion-based quantum computation, Nat. Commun. 14 (2023)
M. McEwen, K.C. Miao, A. Mieszala, S. Montazeri, A. Morvan, R. Movassagh, http://dx.doi.org/10.1038/s41467-023-36493-1.
W. Mruczkiewicz, M. Neeley, C. Neill, A. Nersisyan, M. Newman, J.H. Ng, [71] J.-W. Pan, D. Bouwmeester, H. Weinfurter, A. Zeilinger, Experimental entan-
A. Nguyen, M. Nguyen, M.Y. Niu, T.E. O’Brien, S. Omonije, A. Opremcak, glement swapping: Entangling photons that never interacted, Phys. Rev. Lett.
A. Petukhov, R. Potter, L.P. Pryadko, C. Quintana, C. Rocque, N.C. Rubin, N. 80 (1998) 3892–3894, http://dx.doi.org/10.1103/PhysRevLett.80.3891.
Saei, D. Sank, K. Sankaragomathi, K.J. Satzinger, H.F. Schurkus, C. Schuster, [72] S.L.N. Hermans, M. Pompili, H.K.C. Beukers, S. Baier, J. Borregaard, R. Hanson,
M.J. Shearn, A. Shorter, N. Shutty, V. Shvarts, J. Skruzny, W.C. Smith, R. Qubit teleportation between non-neighbouring nodes in a quantum network,
Somma, G. Sterling, D. Strain, M. Szalay, A. Torres, G. Vidal, B. Villalonga, C.V. Nature 605 (7911) (2022) 663–668, http://dx.doi.org/10.1038/s41586-022-
Heidweiller, T. White, B.W.K. Woo, C. Xing, Z.J. Yao, P. Yeh, J. Yoo, G. Young, 04697-y.

23
D. Barral et al. Computer Science Review 57 (2025) 100747

[73] Y.F. Huang, X.F. Ren, Y.S. Zhang, L.M. Duan, G.C. Guo, Experimental tele- [96] C.T. Nguyen, D.D. Sukachev, M.K. Bhaskar, B. Machielse, D.S. Levonian, E.N.
portation of a quantum controlled-NOT gate, Phys. Rev. Lett. 93 (2004) http: Knall, P. Stroganov, R. Riedinger, H. Park, M. Lončar, M.D. Lukin, Quantum
//dx.doi.org/10.1103/PhysRevLett.93.240501. network nodes based on diamond qubits with an efficient nanophotonic
[74] J.-W. Pan, D. Bouwmeester, M. Daniel, H. Weinfurter, A. Zeilinger, Ex- interface, Phys. Rev. Lett. 123 (2019) http://dx.doi.org/10.1103/PhysRevLett.
perimental test of quantum nonlocality in three-photon Greenberger-Horne- 123.183602.
Zeilinger entanglement, Nature 403 (2000) 515–519, http://dx.doi.org/10. [97] V. Krutyanskiy, M. Galli, V. Krcmarsky, S. Baier, D.A. Fioretto, Y. Pu, A.
1038/35000514. Mazloom, P. Sekatski, M. Canteri, M. Teller, J. Schupp, J. Bate, M. Meraner,
[75] M. Murao, D. Jonathan, M.B. Plenio, V. Vedral, Quantum telecloning and N. Sangouard, B.P. Lanyon, T.E. Northup, Entanglement of trapped-ion qubits
multiparticle entanglement, Phys. Rev. A 59 (1999) 156–161, http://dx.doi. separated by 230 meters, Phys. Rev. Lett. 130 (2023) 050803, http://dx.doi.
org/10.1103/PhysRevA.59.156. org/10.1103/PhysRevLett.130.050803.
[76] Z. Zhao, Y.-A. Chen, A.-N. Zhang, T. Yang, H.J. Briegel, J.-W. Pan, Experimental [98] A.J. Stolk, K.L. van der Enden, M.-C. Slater, I. te Raa-Derckx, P. Botma, J.
demonstration of five-photon entanglement and open-destination teleportation, van Rantwijk, J.J.B. Biemond, R.A.J. Hagen, R.W. Herfst, W.D. Koek, A.J.H.
Nature 430 (6995) (2004) 54–58, http://dx.doi.org/10.1038/nature02643. Meskers, R. Vollmer, E.J. van Zwet, M. Markham, A.M. Edmonds, J.F. Geus,
[77] S.M. Lee, S.W. Lee, H. Jeong, H.S. Park, Quantum teleportation of shared quan- F. Elsen, B. Jungbluth, C. Haefner, C. Tresp, J. Stuhler, S. Ritter, R. Hanson,
tum secret, Phys. Rev. Lett. 124 (2020) http://dx.doi.org/10.1103/PhysRevLett. Metropolitan-scale heralded entanglement of solid-state qubits, Sci. Adv. 10 (44)
124.060501. (2024) eadp6442, http://dx.doi.org/10.1126/sciadv.adp6442.
[78] Z. Zhao, A.N. Zhang, X.Q. Zhou, Y.A. Chen, C.Y. Lu, A. Karlsson, J.W. Pan, [99] J. Ramette, J. Sinclair, Z. Vendeiro, A. Rudelis, M. Cetina, V. Vuletić, Any-
Experimental realization of optimal asymmetric cloning and telecloning via to-any connected cavity-mediated architecture for quantum computing with
partial teleportation, Phys. Rev. Lett. 95 (2005) http://dx.doi.org/10.1103/ trapped ions or rydberg arrays, PRX Quantum 3 (2022) http://dx.doi.org/10.
PhysRevLett.95.030502. 1103/PRXQuantum.3.010344.
[79] L.C. Peng, D. Wu, H.S. Zhong, Y.H. Luo, Y. Li, Y. Hu, X. Jiang, M.C. Chen, L.
[100] Y. Arakawa, M.J. Holmes, Progress in quantum-dot single photon sources for
Li, N.L. Liu, K. Nemoto, W.J. Munro, B.C. Sanders, C.Y. Lu, J.W. Pan, Cloning
quantum information technologies: A broad spectrum overview, Appl. Phys.
of quantum entanglement, Phys. Rev. Lett. 125 (2020) http://dx.doi.org/10.
Rev. 7 (2) (2020) http://dx.doi.org/10.1063/5.0010193.
1103/PhysRevLett.125.210502.
[101] S. Castelletto, A. Boretti, Perspective on solid-state single-photon sources in
[80] Q. Wang, Y. Wang, X. Sun, Y. Tian, W. Li, L. Tian, X. Yu, J. Zhang, Y. Zheng,
the infrared for quantum technology, Adv. Quantum Technol. 6 (10) (2023)
Controllable continuous variable quantum state distributor, Opt. Lett. 46 (2021)
2300145, http://dx.doi.org/10.1002/qute.202300145.
1844, http://dx.doi.org/10.1364/ol.419261.
[81] D. Awschalom, K.K. Berggren, et al., Development of quantum interconnects [102] T. Zhong, P. Goldner, Emerging rare-earth doped material platforms for
(QuICs) for next-generation information technologies, PRX Quantum 2 (2021) quantum nanophotonics, Nanophotonics 8 (11) (2019) 2003–2015, http://dx.
http://dx.doi.org/10.1103/PRXQuantum.2.017002. doi.org/10.1515/nanoph-2019-0185.
[82] H. Edlbauer, J. Wang, T. Crozes, P. Perrier, S. Ouacel, C. Geffroy, G. Georgiou, [103] N. Samkharadze, G. Zheng, N. Kalhor, D. Brousse, A. Sammak, U.C. Mendes,
E. Chatzikyriakou, A. Lacerda-Santos, X. Waintal, D.C. Glattli, P. Roulleau, A. Blais, G. Scappucci, L.M.K. Vandersypen, Strong spin-photon coupling in
J. Nath, M. Kataoka, J. Splettstoesser, M. Acciai, M.C. da Silva Figueira, silicon, Science 359 (6380) (2018) 1123–1127, http://dx.doi.org/10.1126/
K. Öztas, A. Trellakis, T. Grange, O.M. Yevtushenko, S. Birner, C. Bäuerle, science.aar4054.
Semiconductor-based electron flying qubits: Review on recent progress ac- [104] P. Kurpiers, P. Magnard, T. Walter, B. Royer, M. Pechal, J. Heinsoo, Y. Salathé,
celerated by numerical modelling, EPJ Quantum Technol. 9 (1) (2022) 21, A. Akin, S. Storz, J.-C. Besse, S. Gasparinetti, A. Blais, A. Wallraff, Deterministic
http://dx.doi.org/10.1140/epjqt/s40507-022-00139-w. quantum state transfer and remote entanglement using microwave photons,
[83] M. Zhong, M.P. Hedges, R.L. Ahlefeldt, J.G. Bartholomew, S.E. Beavan, S.M. Nature 558 (7709) (2018) 264–267, http://dx.doi.org/10.1038/s41586-018-
Wittig, J.J. Longdell, M.J. Sellars, Optically addressable nuclear spins in a 0195-y.
solid with a six-hour coherence time, Nature 517 (7533) (2015) 177–180, [105] P. Magnard, S. Storz, P. Kurpiers, J. Schär, F. Marxer, J. Lütolf, T. Walter,
http://dx.doi.org/10.1038/nature14025. J.-C. Besse, M. Gabureac, K. Reuer, A. Akin, B. Royer, A. Blais, A. Wallraff,
[84] P. Hurst, A. Miller, Trends in undersea fiber optic systems, in: OCEANS Microwave quantum link between superconducting circuits housed in spatially
2000 MTS/IEEE Conference and Exhibition. Conference Proceedings (Cat. separated cryogenic systems, Phys. Rev. Lett. 125 (2020) 260502, http://dx.
No.00CH37158), Vol. 1, 2000, pp. 479–488, http://dx.doi.org/10.1109/ doi.org/10.1103/PhysRevLett.125.260502.
OCEANS.2000.881303, vol.1. [106] M. Renger, S. Gandorfer, W. Yam, F. Fesquet, M. Handschuh, K.E. Honasoge,
[85] K. Grobe, M. Eiselt, Wavelength Division Multiplexing: A Practical Engineering F. Kronowetter, Y. Nojiri, M. Partanen, M. Pfeiffer, H. van der Vliet, A.J.
Guide, in: Wiley Series in Pure and Applied Optics, Wiley, 2013, http://dx.doi. Matthews, J. Govenius, R.N. Jabdaraghi, M. Prunnila, A. Marx, F. Deppe,
org/10.1002/9781118755068. R. Gross, K.G. Fedorov, Cryogenic microwave link for quantum local area
[86] R.W. Munn, C.N. Ironside, Principles and Applications of Nonlinear Optical networks, 2023, http://dx.doi.org/10.48550/arXiv.2308.12398, arXiv eprint.
Materials, Springer, 1993, http://dx.doi.org/10.1007/978-94-011-2158-3. [107] D. Lago-Rivera, J.V. Rakonjac, S. Grandi, H.d. Riedmatten, Long distance
[87] M. Barbieri, C. Cinelli, P. Mataloni, F. De Martini, Polarization-momentum multiplexed quantum teleportation from a telecom photon to a solid-state
hyperentangled states: Realization and characterization, Phys. Rev. A 72 (2005) qubit, Nat. Commun. 14 (1) (2023) 1889, http://dx.doi.org/10.1038/s41467-
052110, http://dx.doi.org/10.1103/PhysRevA.72.052110. 023-37518-5.
[88] D. Hucul, I.V. Inlek, G. Vittorini, C. Crocker, S. Debnath, S.M. Clark, C. Monroe, [108] P. Zhao, M.-Y. Yang, S. Zhu, L. Zhou, W. Zhong, M.-M. Du, Y.-B. Sheng,
Modular entanglement of atomic qubits using photons and phonons, Nat. Phys. Generation of hyperentangled state encoded in three degrees of freedom, Sci.
11 (1) (2015) 37–42, http://dx.doi.org/10.1038/nphys3150. China Phys. Mech. Astron. 66 (10) (2023) http://dx.doi.org/10.1007/s11433-
[89] D.L. Moehring, P. Maunz, S. Olmschenk, K.C. Younge, D.N. Matsukevich, L.- 023-2164-7.
M. Duan, C. Monroe, Entanglement of single-atom quantum bits at a distance,
[109] W. Li, L. Zhang, H. Tan, Y. Lu, S.-K. Liao, J. Huang, H. Li, Z. Wang, H.-K. Mao,
Nature 449 (7158) (2007) 68–71, http://dx.doi.org/10.1038/nature06118.
B. Yan, Q. Li, Y. Liu, Q. Zhang, C.-Z. Peng, L. You, F. Xu, J.-W. Pan, High-rate
[90] P. Maunz, D.L. Moehring, S. Olmschenk, K.C. Younge, D.N. Matsukevich, C.
quantum key distribution exceeding 110 Mb s–1, Nat. Photonics 17 (5) (2023)
Monroe, Quantum interference of photon pairs from two remote trapped atomic
416–421, http://dx.doi.org/10.1038/s41566-023-01166-4.
ions, Nat. Phys. 3 (8) (2007) 538–541, http://dx.doi.org/10.1038/nphys644.
[110] J. Zhang, E. Zallo, B. Höfer, Y. Chen, R. Keil, M. Zopf, S. Böttner, F. Ding, O.G.
[91] J. Hofmann, M. Krug, N. Ortegel, L. Gérard, M. Weber, W. Rosenfeld, H.
Schmidt, Electric-field-induced energy tuning of on-demand entangled-photon
Weinfurter, Heralded entanglement between widely separated atoms, Science
emission from self-assembled quantum dots, Nano Lett. 17 (1) (2017) 501–507,
337 (6090) (2012) 72–75, http://dx.doi.org/10.1126/science.1221856.
http://dx.doi.org/10.1021/acs.nanolett.6b04539.
[92] E. Togan, Y. Chu, A.S. Trifonov, L. Jiang, J. Maze, L. Childress, M.V.G. Dutt,
A.S. Sørensen, P.R. Hemmer, A.S. Zibrov, M.D. Lukin, Quantum entanglement [111] W. Ou, X. Wang, W. Wei, T. Jin, Y. Zhu, T. Wang, J. Zhang, X. Ou, J.
between an optical photon and a solid-state spin qubit, Nature 466 (7307) Zhang, Strain tuning self-assembled quantum dots for energy-tunable entangled-
(2010) 730–734, http://dx.doi.org/10.1038/nature09256. photon sources using a photolithographically fabricated microelectromechanical
[93] S. Castelletto, A. Edmonds, T. Gaebel, J. Rabeau, Production of multiple system, ACS Photonics 9 (10) (2022) 3421–3428, http://dx.doi.org/10.1021/
diamond-based single-photon sources, IEEE J. Sel. Top. Quantum Electron. 18 acsphotonics.2c01033.
(6) (2012) 1792–1798, http://dx.doi.org/10.1109/JSTQE.2012.2199283. [112] C. Hopfmann, W. Nie, N.L. Sharma, C. Weigelt, F. Ding, O.G. Schmidt, Max-
[94] N. Kalb, A.A. Reiserer, P.C. Humphreys, J.J.W. Bakermans, S.J. Kamerling, imally entangled and gigahertz-clocked on-demand photon pair source, Phys.
N.H. Nickerson, S.C. Benjamin, D.J. Twitchen, M. Markham, R. Hanson, Rev. B 103 (2021) 075413, http://dx.doi.org/10.1103/PhysRevB.103.075413.
Entanglement distillation between solid-state quantum network nodes, Science [113] P. Aumann, M. Prilmüller, F. Kappe, L. Ostermann, D. Dalacu, P.J. Poole,
356 (6341) (2017) 928–932, http://dx.doi.org/10.1126/science.aan0070. H. Ritsch, W. Lechner, G. Weihs, Demonstration and modeling of time-bin
[95] P.C. Humphreys, N. Kalb, J.P.J. Morits, R.N. Schouten, R.F.L. Vermeulen, entangled photons from a quantum dot in a nanowire, AIP Adv. 12 (5) (2022)
D.J. Twitchen, M. Markham, R. Hanson, Deterministic delivery of remote 055115, http://dx.doi.org/10.1063/5.0081874.
entanglement on a quantum network, Nature 558 (7709) (2018) 268–273, [114] L. Zhou, Y.-B. Sheng, Purification of logic-qubit entanglement, Sci. Rep. 6 (1)
http://dx.doi.org/10.1038/s41586-018-0200-5. (2016) http://dx.doi.org/10.1038/srep28813.

24
D. Barral et al. Computer Science Review 57 (2025) 100747

[115] F. Kaiser, P. Vergyris, A. Martin, D. Aktas, M.P.D. Micheli, O. Alibart, S. Tanzilli, [140] Y. Wang, M. Um, J. Zhang, S. An, M. Lyu, J.-N. Zhang, L.-M. Duan, D. Yum,
Quantum optical frequency up-conversion for polarisation entangled qubits: K. Kim, Single-qubit quantum memory exceeding ten-minute coherence time,
Towards interconnected quantum information devices, Opt. Express 27 (18) Nat. Photonics 11 (10) (2017) 646–650, http://dx.doi.org/10.1038/s41566-017-
(2019) 25603–25610, http://dx.doi.org/10.1364/OE.27.025603. 0007-1.
[116] S. Murakami, R. Fujimoto, T. Kobayashi, R. Ikuta, A. Inoue, T. Umeki, S. Miki, [141] ITU-T, Transmission Media and Optical Systems Characteristics – Optical Fibre
F. China, H. Terai, R. Kasahara, T. Mukai, N. Imoto, T. Yamamoto, Quantum Cables, Standard, Telecomm. Standardization Sector of ITU, 2016.
frequency conversion using 4-port fiber-pigtailed PPLN module, Opt. Express [142] Y. Tamura, H. Sakuma, K. Morita, M. Suzuki, Y. Yamamoto, K. Shimada, Y.
31 (18) (2023) 29271–29279, http://dx.doi.org/10.1364/OE.494313. Honma, K. Sohma, T. Fujii, T. Hasegawa, The first 0.14-dB/km loss optical
[117] M.J. Weaver, P. Duivestein, A.C. Bernasconi, S. Scharmer, M. Lemang, T.C.v. fiber and its impact on submarine transmission, J. Lightwave Technol. 36 (1)
Thiel, F. Hijazi, B. Hensen, S. Gröblacher, R. Stockill, An integrated microwave- (2018) 44–49, http://dx.doi.org/10.1109/JLT.2018.2796647.
to-optics interface for scalable quantum computing, Nature Nanotechnology 19 [143] E. Sugita, R. Nagase, K. Kanayama, T. Shintaku, SC-type single-mode optical
(2) (2023) 166–172, http://dx.doi.org/10.1038/s41565-023-01515-y. fiber connectors, J. Lightwave Technol. 7 (11) (1989) 1689–1696, http://dx.
[118] R. Sahu, L. Qiu, W. Hease, G. Arnold, Y. Minoguchi, P. Rabl, J.M. Fink, doi.org/10.1109/50.45890.
Entangling microwaves with light, Science 380 (6646) (2023) 718–721, http: [144] Y. Yamamoto, The quantum optical repeater, Science 263 (5152) (1994)
//dx.doi.org/10.1126/science.adg3812. 1394–1395, http://dx.doi.org/10.1126/science.263.5152.1394.
[119] T.B. Pittman, B.C. Jacobs, J.D. Franson, Single photons on pseudodemand
[145] M. Takeoka, S. Guha, M.M. Wilde, Fundamental rate-loss tradeoff for optical
from stored parametric down-conversion, Phys. Rev. A 66 (2002) 042303,
quantum key distribution, Nat. Commun. 5 (1) (2014) 5235, http://dx.doi.org/
http://dx.doi.org/10.1103/PhysRevA.66.042303.
10.1038/ncomms6235.
[120] O. Landry, J.A.W. van Houwelingen, A. Beveratos, H. Zbinden, N. Gisin, Quan-
[146] J. Miguel-Ramiro, W. Dür, Efficient entanglement purification protocols for 𝑑-
tum teleportation over the Swisscom telecommunication network, J. Opt. Soc.
level systems, Phys. Rev. A 98 (2018) http://dx.doi.org/10.1103/PhysRevA.98.
Am. B 24 (2) (2007) 398–403, http://dx.doi.org/10.1364/JOSAB.24.000398.
042309.
[121] P.M. Leung, T.C. Ralph, Quantum memory scheme based on optical fibers
[147] X.-M. Hu, C.-X. Huang, Y.-B. Sheng, L. Zhou, B.-H. Liu, Y. Guo, C. Zhang, W.-B.
and cavities, Phys. Rev. A 74 (2006) http://dx.doi.org/10.1103/PhysRevA.74.
Xing, Y.-F. Huang, C.-F. Li, G.-C. Guo, Long-distance entanglement purification
022311.
for quantum communication, Phys. Rev. Lett. 126 (2021) http://dx.doi.org/10.
[122] T. Tanabe, M. Notomi, E. Kuramochi, A. Shinya, H. Taniyama, Trapping and
1103/PhysRevLett.126.010503.
delaying photons for one nanosecond in an ultrasmall high-Q photonic-crystal
nanocavity, Nat. Photonics 1 (1) (2007) 49–52, http://dx.doi.org/10.1038/ [148] F.-F. Du, G. Fan, Y.-M. Wu, B.-C. Ren, Faithful and efficient hyperentanglement
nphoton.2006.51. purification for spatial-polarization-time-bin photon system, Chin. Phys. B 32
[123] S.A. Moiseev, B.S. Ham, Photon-echo quantum memory with efficient multipulse (6) (2023) http://dx.doi.org/10.1088/1674-1056/aca395.
readings, Phys. Rev. A 70 (2004) http://dx.doi.org/10.1103/PhysRevA.70. [149] C.-C. Luo, L. Zhou, W. Zhong, Y.-B. Sheng, Purification for hybrid logical qubit
063809. entanglement, Quantum Inf. Process. 21 (8) (2022) 300, http://dx.doi.org/10.
[124] A.I. Lvovsky, B.C. Sanders, W. Tittel, Optical quantum memory, Nat. Photonics 1007/s11128-022-03646-y.
3 (12) (2009) 706–714, http://dx.doi.org/10.1038/nphoton.2009.231. [150] T. van Leent, M. Bock, F. Fertig, R. Garthoff, S. Eppelt, Y. Zhou, P. Malik, M.
[125] M. Guo, S. Liu, W. Sun, M. Ren, F. Wang, M. Zhong, Rare-earth quantum Seubert, T. Bauer, W. Rosenfeld, W. Zhang, C. Becher, H. Weinfurter, Entangling
memories: The experimental status quo, Front. Phys. 18 (2) (2023) 21303, single atoms over 33 km telecom fibre, Nature 607 (7917) (2022) 69–73,
http://dx.doi.org/10.1007/s11467-022-1240-8. http://dx.doi.org/10.1038/s41586-022-04764-4.
[126] A.V. Gorshkov, A. André, M. Fleischhauer, A.S. Sørensen, M.D. Lukin, Universal [151] S. Ourari, Ł. Dusanowski, S.P. Horvath, M.T. Uysal, C.M. Phenicie, P. Stevenson,
approach to optimal photon storage in atomic media, Phys. Rev. Lett. 98 (2007) M. Raha, S. Chen, R.J. Cava, N.P. de Leon, J.D. Thompson, Indistinguishable
123601, http://dx.doi.org/10.1103/PhysRevLett.98.123601. telecom band photons from a single Er ion in the solid state, Nature 620 (7976)
[127] W. Tittel, M. Afzelius, T. Chaneliére, R. Cone, S. Kröll, S. Moiseev, M. Sellars, (2023) 977–981, http://dx.doi.org/10.1038/s41586-023-06281-4.
Photon-echo quantum memory in solid state systems, Laser Photonics Rev. 4 [152] J. Laurat, On-demand entanglement could lead to scalable quantum networks,
(2) (2010) 244–267, http://dx.doi.org/10.1002/lpor.200810056. Nature 558 (7709) (2018) 192–193, http://dx.doi.org/10.1038/d41586-018-
[128] M. Afzelius, C. Simon, H. de Riedmatten, N. Gisin, Multimode quantum memory 05336-1.
based on atomic frequency combs, Phys. Rev. A 79 (2009) http://dx.doi.org/ [153] Q. Zhu, Y. Zhao, H. Xu, L. Huang, C. Qiao, Integrating all-optical switching and
10.1103/PhysRevA.79.052329. entangled photon source placement for entanglement routing, in: IEEE/ACM
[129] A. Ortu, A. Holzäpfel, J. Etesse, M. Afzelius, Storage of photonic time-bin qubits 31st Int. Symp. on Quality of Service, IWQoS, 2023, pp. 1–10, http://dx.doi.
for up to 20 ms in a rare-earth doped crystal, Npj Quantum Inf. 8 (1) (2022) org/10.1109/IWQoS57198.2023.10188776.
29, http://dx.doi.org/10.1038/s41534-022-00541-3. [154] A.S. Cacciapuoti, M. Viscardi, J. Illiano, M. Caleffi, Entanglement distribution
[130] E. Knill, R. Laflamme, G.J. Milburn, A scheme for efficient quantum computa- in the quantum internet: Knowing when to stop!, 2023, http://dx.doi.org/10.
tion with linear optics, Nature 409 (6816) (2001) 46–52, http://dx.doi.org/10. 48550/arXiv.2307.05123, arxiv eprint.
1038/35051009. [155] Y. Zeng, J. Zhang, J. Liu, Z. Liu, Y. Yang, Entanglement routing design over
[131] P. Kok, W.J. Munro, K. Nemoto, T.C. Ralph, J.P. Dowling, G.J. Milburn, Linear quantum networks, IEEE/ACM Trans. Netw. 32 (1) (2023) 352–367, http:
optical quantum computing with photonic qubits, Rev. Modern Phys. 79 (2007) //dx.doi.org/10.1109/TNET.2023.3282560.
135–174, http://dx.doi.org/10.1103/RevModPhys.79.135. [156] L. Gyongyosi, S. Imre, Advances in the quantum internet, Commun. ACM 65
[132] E. Knill, R. Laflamme, W.H. Zurek, Resilient quantum computation, Science 279 (8) (2022) 52–63, http://dx.doi.org/10.1145/3524455.
(5349) (1998) 342–345, http://dx.doi.org/10.1126/science.279.5349.342.
[157] H.J. Kimble, The quantum internet, Nature 453 (7198) (2008) 1023–1030,
[133] A.Y. Kitaev, Fault-tolerant quantum computation by anyons, Ann. Physics 303
http://dx.doi.org/10.1038/nature07127.
(1) (2003) 2–30, http://dx.doi.org/10.1016/S0003-4916(02)00018-0.
[158] S. Wehner, D. Elkouss, R. Hanson, Quantum internet: A vision for the
[134] T.P. Harty, D.T.C. Allcock, C.J. Ballance, L. Guidoni, H.A. Janacek, N.M. Linke,
road ahead, Science 362 (6412) (2018) eaam9288, http://dx.doi.org/10.1126/
D.N. Stacey, D.M. Lucas, High-fidelity preparation, gates, memory, and readout
science.aam9288.
of a trapped-ion quantum bit, Phys. Rev. Lett. 113 (2014) http://dx.doi.org/
[159] A.S. Cacciapuoti, M. Caleffi, F. Tafuri, F.S. Cataliotti, S. Gherardini, G. Bianchi,
10.1103/PhysRevLett.113.220501.
Quantum internet: Networking challenges in distributed quantum computing,
[135] N. Schlosser, G. Reymond, I. Protsenko, P. Grangier, Sub-poissonian loading
IEEE Netw. 34 (1) (2020) 137–143, http://dx.doi.org/10.1109/MNET.001.
of single atoms in a microscopic dipole trap, Nature 411 (6841) (2001)
1900092.
1024–1027, http://dx.doi.org/10.1038/35082512.
[136] M.A. Norcia, A.W. Young, A.M. Kaufman, Microscopic control and detection of [160] M. Pant, H. Krovi, D. Towsley, L. Tassiulas, L. Jiang, P. Basu, D. Englund, S.
ultracold strontium in optical-tweezer arrays, Phys. Rev. X 8 (2018) 041054, Guha, Routing entanglement in the quantum internet, Npj Quantum Inf. 5 (1)
http://dx.doi.org/10.1103/PhysRevX.8.041054. (2019) 25, http://dx.doi.org/10.1038/s41534-019-0139-x.
[137] K. Barnes, P. Battaglino, B.J. Bloom, K. Cassella, R. Coxe, N. Crisosto, J.P. [161] A. Reiserer, G. Rempe, Cavity-based quantum networks with single atoms and
King, S.S. Kondov, K. Kotru, S.C. Larsen, J. Lauigan, B.J. Lester, M. McDonald, optical photons, Rev. Modern Phys. 87 (2015) 1379–1418, http://dx.doi.org/
E. Megidish, S. Narayanaswami, C. Nishiguchi, R. Notermans, L.S. Peng, A. 10.1103/RevModPhys.87.1379.
Ryou, T.-Y. Wu, M. Yarwood, Assembly and coherent control of a register of [162] Z. Li, K. Xue, J. Li, L. Chen, R. Li, Z. Wang, N. Yu, D.S. Wei, Q. Sun, J.
nuclear spin qubits, Nat. Commun. 13 (1) (2022) 2779, http://dx.doi.org/10. Lu, Entanglement-assisted quantum networks: Mechanics, enabling technologies,
1038/s41467-022-29977-z. challenges, and research directions, IEEE Commun. Surv. Tutor. 25 (4) (2023)
[138] L. Isenhower, E. Urban, X.L. Zhang, A.T. Gill, T. Henage, T.A. Johnson, T.G. 2133–2189, http://dx.doi.org/10.1109/COMST.2023.3294240.
Walker, M. Saffman, Demonstration of a neutral atom controlled-NOT quantum [163] J. Miguel-Ramiro, A. Pirker, W. Dür, Genuine quantum networks with super-
gate, Phys. Rev. Lett. 104 (2010) http://dx.doi.org/10.1103/PhysRevLett.104. posed tasks and addressing, Npj Quantum Inf. 7 (2021) http://dx.doi.org/10.
010503. 1038/s41534-021-00472-5.
[139] M.W. Doherty, N.B. Manson, P. Delaney, F. Jelezko, J. Wrachtrup, L.C. [164] J. Illiano, M. Caleffi, A. Manzalini, A.S. Cacciapuoti, Quantum internet protocol
Hollenberg, The nitrogen-vacancy colour centre in diamond, Phys. Rep. 528 stack: A comprehensive survey, Comput. Netw. 213 (2022) 109092, http:
(1) (2013) 1–45, http://dx.doi.org/10.1016/j.physrep.2013.02.001. //dx.doi.org/10.1016/j.comnet.2022.109092.

25
D. Barral et al. Computer Science Review 57 (2025) 100747

[165] W. Kozlowski, S. Wehner, R.V. Meter, B. Rijsman, A.S. Cacciapuoti, M. Caleffi, [191] S. Schlag, T. Heuer, L. Gottesbüren, Y. Akhremtsev, C. Schulz, P. Sanders, High-
S. Nagayama, RFC 9340: Architectural principles for a quantum internet, 2023, quality hypergraph partitioning, ACM J. Exp. Algorithmics 27 (2023) 1–39,
http://dx.doi.org/10.17487/RFC9340. http://dx.doi.org/10.1145/3529090.
[166] M. Azari, P. Polakos, K.P. Seshadreesan, Quantum switches for Gottesman– [192] M. Houshmand, Z. Mohammadi, M. Zomorodi-Moghadam, M. Houshmand, An
Kitaev–Preskill qubit-based all-photonic quantum networks, 2024, http://dx.doi. evolutionary approach to optimizing teleportation cost in distributed quantum
org/10.48550/arXiv.2402.02721, arxiv preprint. computation, Int. J. Theor. Phys. 59 (4) (2020) 1315–1329, http://dx.doi.org/
[167] I. Tillman, T. Vasantam, K.P. Seshadreesan, A continuous variable quantum 10.1007/S10773-020-04409-0.
switch, in: Proc. of Int. Conf. on Quantum Computing and Engineering, QCE, [193] O. Daei, K. Navi, M. Zomorodi-Moghadam, Optimized quantum circuit parti-
IEEE, 2022, pp. 365–371, http://dx.doi.org/10.1109/QCE53715.2022.00057. tioning, Internat. J. Theoret. Phys. 59 (12) (2020) 3804–3820, http://dx.doi.
[168] T.C. Ralph, A.P. Lund, Nondeterministic Noiseless Linear Amplification of org/10.1007/s10773-020-04633-8.
Quantum Systems, Vol. 1110, 2009, pp. 155–160, http://dx.doi.org/10.1063/ [194] E. Nikahd, N. Mohammadzadeh, M. Sedighi, M.S. Zamani, Automated window-
1.3131295. based partitioning of quantum circuits, Phys. Scr. 96 (3) (2021) 035102,
[169] A. Pirker, W. Dür, A quantum network stack and protocols for reliable http://dx.doi.org/10.1088/1402-4896/ABD57C.
entanglement-based networks, New J. Phys. 21 (3) (2019) http://dx.doi.org/ [195] J.-Y. Wu, K. Matsui, T. Forrer, A. Soeda, P. Andrés-Martínez, D. Mills, L. Henaut,
10.1088/1367-2630/ab05f7. M. Murao, Entanglement-efficient bipartite-distributed quantum computing,
[170] R.V. Meter, J. Touch, Designing quantum repeater networks, IEEE Commun. Quantum 7 (2023) 1196, http://dx.doi.org/10.22331/q-2023-12-05-1196.
Mag. 51 (8) (2013) 64–71, http://dx.doi.org/10.1109/MCOM.2013.6576340. [196] P. Andrés-Martínez, T. Forrer, D. Mills, J.-Y. Wu, L. Henaut, K. Yamamoto,
[171] R. Van Meter, R. Satoh, N. Benchasattabuse, K. Teramoto, T. Matsuo, M. M. Murao, R. Duncan, Distributing circuits over heterogeneous, modular quan-
Hajdušek, T. Satoh, S. Nagayama, S. Suzuki, A quantum internet architecture, tum computing network architectures, 2023, http://dx.doi.org/10.48550/arXiv.
in: 2022 IEEE Int. Conf. on Quantum Computing and Engineering, QCE, 2022, 2305.14148, arxiv preprint.
pp. 341–352, http://dx.doi.org/10.1109/QCE53715.2022.00055. [197] R. G. Sundaram, H. Gupta, C.R. Ramakrishnan, Efficient distribution of quantum
[172] Z. Li, K. Xue, J. Li, N. Yu, J. Liu, D.S.L. Wei, Q. Sun, J. Lu, Building a large-scale circuits, in: 35th Int. Symp. on Distributed Computing, DISC, in: Leibniz
and wide-area quantum internet based on an OSI-alike model, China Commun. International Proceedings in Informatics (LIPIcs), vol. 209, 2021, pp. 1–20,
18 (10) (2021) 1–14, http://dx.doi.org/10.23919/JCC.2021.10.001. http://dx.doi.org/10.4230/LIPIcs.DISC.2021.41.
[173] A. Dahlberg, M. Skrzypczyk, T. Coopmans, L. Wubben, F. Rozpundefineddek, [198] R.G. Sundaram, H. Gupta, C.R. Ramakrishnan, Distribution of quantum circuits
M. Pompili, A. Stolk, P. Pawełczak, R. Knegjens, J. de Oliveira Filho, R. Hanson, over general quantum networks, in: 2022 IEEE Int. Conf. on Quantum Com-
S. Wehner, A link layer protocol for quantum networks, in: Proceedings of the puting and Engineering, QCE, 2022, pp. 415–425, http://dx.doi.org/10.1109/
ACM Special Interest Group on Data Communication, ACM, New York, NY, QCE53715.2022.00063.
USA, 2019, pp. 159–173, http://dx.doi.org/10.1145/3341302.3342070. [199] R.G. Sundaram, H. Gupta, Distributing quantum circuits using teleportations,
[174] A.S. Cacciapuoti, J. Illiano, M. Caleffi, Quantum internet addressing, IEEE Netw. in: 2023 IEEE Int. Conf. on Quantum Software, QSW, 2023, pp. 186–192,
38 (1) (2024) 104–111, http://dx.doi.org/10.1109/MNET.2023.3328393. http://dx.doi.org/10.1109/QSW59989.2023.00030.
[175] J. Illiano, M. Caleffi, M. Viscardi, A.S. Cacciapuoti, Quantum MAC: Genuine
[200] Z. Davarzani, M. Zomorodi-Moghadam, M. Houshmand, M. Nouri-Baygi, A
entanglement access control via many-body dicke states, IEEE Trans. Commun.
dynamic programming approach for distributing quantum circuits by bipartite
72 (4) (2024) 2090–2105, http://dx.doi.org/10.1109/TCOMM.2023.3344140.
graphs, Quantum Inf. Process. 19 (2020) http://dx.doi.org/10.1007/s11128-
[176] M. Caleffi, Optimal routing for quantum networks, IEEE Access 5 (2017)
020-02871-7.
22299–22312, http://dx.doi.org/10.1109/ACCESS.2017.2763325.
[201] J. Clark, T. Humble, H. Thapliyal, TDAG: Tree-based directed acyclic graph
[177] N. Yu, C.-Y. Lai, L. Zhou, Protocols for packet quantum network intercommu-
partitioning for quantum circuits, in: Proc. of the Great Lakes Symposium on
nication, IEEE Trans. Quantum Eng. 2 (2021) 1–9, http://dx.doi.org/10.1109/
VLSI, ACM, 2023, pp. 587–592, http://dx.doi.org/10.1145/3583781.3590234.
TQE.2021.3112594.
[202] W. Cambiucci, R. Silveira, W. Ruggiero, Hypergraphic partitioning of quan-
[178] Defense Advanced Research Projects Agency, Transmission control protocol,
tum circuits for distributed quantum computing, in: 2023 IEEE Int. Conf.
1981, URL https://www.ietf.org/rfc/rfc793.txt. (Accessed 21 January 2025).
on Quantum Computing and Engineering, QCE, IEEE Computer Society, Los
[179] R. Cleve, D. Gottesman, H.-K. Lo, How to share a quantum secret, Phys. Rev.
Alamitos, CA, USA, 2023, pp. 268–269, http://dx.doi.org/10.1109/QCE57702.
Lett. 83 (1999) 648–651, http://dx.doi.org/10.1103/PhysRevLett.83.648.
2023.10237.
[180] C.D. Donne, M. Iuliano, B. van der Vecht, G.M. Ferreira, H. Jirovská, T.
[203] C.M. Fiduccia, R.M. Mattheyses, A linear-time heuristic for improving network
van der Steenhoven, A. Dahlberg, M. Skrzypczyk, D. Fioretto, M. Teller, P.
partitions, in: Papers on Twenty-Five Years of Electronic Design Automation,
Filippov, A.R.-P. Montblanch, J. Fischer, B. van Ommen, N. Demetriou, D.
1982, pp. 175–181, http://dx.doi.org/10.1109/DAC.1982.1585498.
Leichtle, L. Music, H. Ollivier, I. te Raa, W. Kozlowski, T. Taminiau, P.
[204] J.M. Baker, C. Duckering, A. Hoover, F.T. Chong, Time-sliced quantum circuit
Pawełczak, T. Northup, R. Hanson, S. Wehner, Design and demonstration of an
partitioning for modular architectures, in: Proc. of the 17th ACM Int. Conf.
operating system for executing applications on quantum network nodes, 2024,
on Computing Frontiers, ACM, 2020, pp. 98–107, http://dx.doi.org/10.1145/
http://dx.doi.org/10.48550/arXiv.2407.18306.
[181] T. Satoh, R.V. Meter, Quantum sockets for the NISQ era quantum 3387902.3392617.
internet, 2018, URL https://www.ngc.is.ritsumei.ac.jp/~ger/static/AQIS18/ [205] T. Park, C.Y. Lee, Algorithms for partitioning a graph, Comput. Ind. Eng. 28
OnlineBooklet/158.pdf. (Accessed 21 January 2025). (4) (1995) 899–909, http://dx.doi.org/10.1016/0360-8352(95)00003-J.
[182] S. Gauthier, G. Vardoyan, S. Wehner, A control architecture for entanglement [206] A. Ovide, S. Rodrigo, M. Bandic, H. Van Someren, S. Feld, S. Abadal, E.
generation switches in quantum networks, in: Proc. of the 1st Workshop on Alarcon, C.G. Almudever, Mapping quantum algorithms to multi-core quantum
Quantum Networks and Distributed Quantum Computing, QuNet, ACM, 2023, computing architectures, in: IEEE Int. Symp. on Circuits and Systems, ISCAS,
pp. 38–44, http://dx.doi.org/10.1145/3610251.3610552. IEEE, 2023, pp. 1–5, http://dx.doi.org/10.1109/ISCAS46773.2023.10181589.
[183] Z. Xiao, J. Li, K. Xue, Z. Li, N. Yu, Q. Sun, J. Lu, A connectionless entanglement [207] P. Escofet, A. Ovide, C.G. Almudever, E. Alarcón, S. Abadal, Hungarian
distribution protocol design in quantum networks, IEEE Netw. 38 (1) (2024) qubit assignment for optimized mapping of quantum circuits on multi-core
131–139, http://dx.doi.org/10.1109/MNET.2023.3321044. architectures, IEEE Comput. Archit. Lett. (2023) http://dx.doi.org/10.1109/
[184] C. Amza, A.L. Cox, S. Dwarkadas, P. Keleher, H. Lu, R. Rajamony, W. LCA.2023.3318857.
Yu, W. Zwaenepoel, TreadMarks: Shared memory computing on networks of [208] M. Bandic, L. Prielinger, J. Nüßlein, A. Ovide, S. Rodrigo, S. Abadal, H. van
workstations, Computer 29 (1996) 18–28, http://dx.doi.org/10.1109/2.485843. Someren, G. Vardoyan, E. Alarcon, C.G. Almudever, et al., Mapping quantum
[185] M. Van Steen, A.S. Tanenbaum, Distributed Systems, Maarten van Steen Leiden, circuits to modular architectures with QUBO, in: 2023 IEEE Int. Conf. on
The Netherlands, 2017, URL https://www.distributed-systems.net/index.php/ Quantum Computing and Engineering, QCE, Vol. 1, IEEE, 2023, pp. 790–801,
books/ds3/. http://dx.doi.org/10.1109/QCE57702.2023.00094.
[186] C. Berge, Hypergraphs: Combinatorics of Finite Sets, in: North-Holland Mathe- [209] A. Pastor, P. Escofet, S.B. Rached, E. Alarcón, P. Barlet-Ros, S. Abadal, Circuit
matical Library, Elsevier Science, 1984, URL https://shop.elsevier.com/books/ partitioning for multi-core quantum architectures with deep reinforcement
hypergraphs/berge/978-0-444-87489-4. learning, 2024, http://dx.doi.org/10.48550/arXiv.2401.17976, arxiv preprint.
[187] M. Zomorodi-Moghadam, M. Houshmand, M. Houshmand, Optimizing telepor- [210] H. Pashayan, J.J. Wallman, S.D. Bartlett, Estimating outcome probabilities
tation cost in distributed quantum circuits, Int. J. Theor. Phys. 57 (3) (2018) of quantum circuits using quasiprobabilities, Phys. Rev. Lett. 115 (7) (2015)
848–861, http://dx.doi.org/10.1007/S10773-017-3618-X. http://dx.doi.org/10.1103/PhysRevLett.115.070501.
[188] B.W. Kernighan, S. Lin, An efficient heuristic procedure for partitioning graphs, [211] K. Mitarai, K. Fujii, Overhead for simulating a non-local channel with local
Bell Syst. Tech. J. 49 (2) (1970) 291–307, http://dx.doi.org/10.1002/j.1538- channels by quasiprobability sampling, Quantum 5 (2021) 388, http://dx.doi.
7305.1970.tb01770.x. org/10.22331/q-2021-01-28-388.
[189] P. Andrés-Martínez, C. Heunen, Automated distribution of quantum circuits via [212] C. Piveteau, D. Sutter, S. Woerner, Quasiprobability decompositions with
hypergraph partitioning, Phys. Rev. A 100 (3) (2019) 032308, http://dx.doi. reduced sampling overhead, Npj Quantum Inf. 8 (1) (2022) 1–9, http://dx.
org/10.1103/PhysRevA.100.032308. doi.org/10.1038/s41534-022-00517-3.
[190] Y. Akhremtsev, T. Heuer, P. Sanders, S. Schlag, Engineering a direct k-way [213] H.F. Hofmann, How to simulate a universal quantum computer using negative
hypergraph partitioning algorithm, in: Proc. of ALENEX, SIAM, 2017, pp. 28–42, probabilities, J. Phys. A 42 (27) (2009) 275304, http://dx.doi.org/10.1088/
http://dx.doi.org/10.1137/1.9781611974768.3. 1751-8113/42/27/275304.

26
D. Barral et al. Computer Science Review 57 (2025) 100747

[214] S. Bravyi, G. Smith, J.A. Smolin, Trading classical and quantum computational [239] Y. Zhang, L. Cincio, C.F.A. Negre, P. Czarnik, P.J. Coles, P.M. Anisimov, S.M.
resources, Phys. Rev. X 6 (2) (2016) http://dx.doi.org/10.1103/PhysRevX.6. Mniszewski, S. Tretiak, P.A. Dub, Variational quantum eigensolver with reduced
021043. circuit complexity, Npj Quantum Inf. 8 (96) (2022) 1–10, http://dx.doi.org/10.
[215] K. Temme, S. Bravyi, J.M. Gambetta, Error mitigation for short-depth quan- 1038/s41534-022-00599-z.
tum circuits, Phys. Rev. Lett. 119 (18) (2017) http://dx.doi.org/10.1103/ [240] S. Zhang, Z. Qin, Y. Zhou, R. Li, C. Du, Z. Xiao, Single entanglement connection
PhysRevLett.119.180509. architecture between multi-layer HEA for distributed VQE, 2023, http://dx.doi.
[216] S. Endo, S.C. Benjamin, Y. Li, Practical quantum error mitigation for near-future org/10.48550/arXiv.2307.12323, arxiv preprint.
applications, Phys. Rev. X 8 (3) (2018) http://dx.doi.org/10.1103/PhysRevX.8. [241] G. Gentinetta, F. Metz, G. Carleo, Overhead-constrained circuit knitting for
031027. variational quantum dynamics, Quantum 8 (2024) 1296, http://dx.doi.org/10.
[217] C. Ying, B. Cheng, Y. Zhao, H.-L. Huang, Y.-N. Zhang, M. Gong, Y. Wu, S. 22331/q-2024-03-21-1296.
[242] R. Nagai, S. Kanno, Y. Sato, N. Yamamoto, Quantum channel decomposition
Wang, F. Liang, J. Lin, Y. Xu, H. Deng, H. Rong, C.-Z. Peng, M.-H. Yung,
with preselection and postselection, Phys. Rev. A 108 (2) (2023) http://dx.doi.
X. Zhu, J.-W. Pan, Experimental simulation of larger quantum circuits with
org/10.1103/PhysRevA.108.022615.
fewer superconducting qubits, Phys. Rev. Lett. 130 (11) (2023) 110601, http:
[243] D. Chen, B. Baheri, V. Chaudhary, Q. Guan, N. Xie, S. Xu, Approximate
//dx.doi.org/10.1103/PhysRevLett.130.110601.
quantum circuit cutting, 2022, http://dx.doi.org/10.48550/arXiv.2212.01270,
[218] A.P. Singh, K. Mitarai, Y. Suzuki, K. Heya, Y. Tabuchi, K. Fujii, Y. Nakamura,
arxiv preprint.
Experimental demonstration of a high-fidelity virtual two-qubit gate, Phys. Rev.
[244] D.T. Chen, E.H. Hansen, X. Li, V. Kulkarni, V. Chaudhary, B. Ren, Q. Guan,
Res. 6 (2024) http://dx.doi.org/10.1103/PhysRevResearch.6.013235.
S. Kuppannagari, J. Liu, S. Xu, Efficient quantum circuit cutting by neglecting
[219] J.A. Smolin, J.M. Gambetta, G. Smith, Efficient method for computing the basis elements, in: IEEE IPDPS Workshops, IPDPSW, 2023, pp. 517–523, http:
maximum-likelihood quantum state from measurements with additive Gaussian //dx.doi.org/10.1109/IPDPSW59300.2023.00091.
noise, Phys. Rev. Lett. 108 (7) (2012) 070502, http://dx.doi.org/10.1103/ [245] S. Brandhofer, I. Polian, K. Krsulich, Optimal partitioning of quantum circuits
PhysRevLett.108.070502. using gate cuts and wire cuts, IEEE Trans. Quantum Eng. 5 (2024) 1–10,
[220] M.A. Perlin, Z.H. Saleem, M. Suchara, J.C. Osborn, Quantum circuit cutting http://dx.doi.org/10.1109/TQE.2023.3347106.
with maximum-likelihood tomography, Npj Quantum Inf. 7 (64) (2021) 1–8, [246] W. Tang, T. Tomesh, M. Suchara, J. Larson, M. Martonosi, CutQC: Using small
http://dx.doi.org/10.1038/s41534-021-00390-6. quantum computers for large quantum circuit evaluations, in: Proc. of the
[221] A. Eddins, M. Motta, T.P. Gujarati, S. Bravyi, A. Mezzacapo, C. Hadfield, S. 26th ACM Int. Conf. on Architectural Support for Programming Languages
Sheldon, Doubling the size of quantum simulators by entanglement forging, and Operating Systems, ACM, 2021, pp. 473–486, http://dx.doi.org/10.1145/
PRX Quantum 3 (1) (2022) 010309, http://dx.doi.org/10.1103/PRXQuantum. 3445814.3446758.
3.010309. [247] W. Tang, M. Martonosi, ScaleQC: A scalable framework for hybrid computation
[222] D. Faílde, J. Santos-Suárez, D.A. Herrera-Martí, J. Mas, Hamiltonian forging on quantum and classical processors, 2022, http://dx.doi.org/10.48550/arXiv.
of a thermofield double, 2023, http://dx.doi.org/10.48550/arXiv.2311.10566, 2207.00933, arxiv preprint.
arxiv preprint. [248] K.N. Smith, M.A. Perlin, P. Gokhale, P. Frederick, D. Owusu-Antwi, R. Rines,
[223] K. Mitarai, K. Fujii, Constructing a virtual two-qubit gate by sampling single- V. Omole, F. Chong, Clifford-based circuit cutting for quantum simulation, in:
qubit operations, New J. Phys. 23 (2) (2021) 023021, http://dx.doi.org/10. Proc. of the 50th Int. Symp. on Computer Architecture, ACM, 2023, pp. 1–13,
1088/1367-2630/abd7bc. http://dx.doi.org/10.1145/3579371.3589352.
[224] T. Peng, A.W. Harrow, M. Ozols, X. Wu, Simulating large quantum circuits on [249] T. Chatterjee, A. Das, S.I. Mohtashim, A. Saha, A. Chakrabarti, Qurzon: A
a small quantum computer, Phys. Rev. Lett. 125 (15) (2020) http://dx.doi.org/ prototype for a divide and conquer-based quantum compiler for distributed
10.1103/PhysRevLett.125.150504. quantum systems, SN Comput. Sci. 3 (4) (2022) http://dx.doi.org/10.1007/
[225] L. Brenner, C. Piveteau, D. Sutter, Optimal wire cutting with classi- s42979-022-01207-9.
[250] L.S. Herzog, F. Wagner, C. Ufrecht, L. Palackal, A. Plinge, C. Mutschler, D.D.
cal communication, 2023, http://dx.doi.org/10.48550/arXiv.2302.03366, arxiv
Scherer, Improving Quantum and Classical Decomposition Methods for Vehicle
preprint.
Routing, 2024, http://dx.doi.org/10.48550/arXiv.2404.05551, arXiv.
[226] A. Peruzzo, J. McClean, P. Shadbolt, M.-H. Yung, X.-Q. Zhou, P.J. Love, A.
[251] A. Robert, P.K. Barkoutsos, S. Woerner, I. Tavernelli, Resource-efficient quan-
Aspuru-Guzik, J.L. O’Brien, A variational eigenvalue solver on a photonic
tum algorithm for protein folding, Npj Quantum Inf. 7 (1) (2021) 38, http:
quantum processor, Nat. Commun. 5 (1) (2014) 4213, http://dx.doi.org/10.
//dx.doi.org/10.1038/s41534-021-00368-4.
1038/ncomms5213.
[252] J. Tilly, H. Chen, S. Cao, D. Picozzi, K. Setia, Y. Li, E. Grant, L. Wossnig,
[227] S.C. Marshall, J. Tura, V. Dunjko, All this for one qubit? Bounds on local I. Rungger, G.H. Booth, J. Tennyson, The variational quantum eigensolver: A
circuit cutting schemes, 2023, http://dx.doi.org/10.48550/arXiv.2303.13422, review of methods and best practices, Phys. Rep. 986 (2022) 1–128, http://dx.
arxiv preprint, arXiv:2303.13422. doi.org/10.1016/j.physrep.2022.08.003, The Variational Quantum Eigensolver:
[228] C. Piveteau, D. Sutter, Circuit knitting with classical communication, IEEE A Review of Methods and Best Practices.
Trans. Inform. Theory (2023) http://dx.doi.org/10.1109/TIT.2023.3310797, [253] Y. Wang, L.M. Sager-Smith, D.A. Mazziotti, Quantum simulation of bosons
1–1. with the contracted quantum eigensolver, New J. Phys. 25 (10) (2023) http:
[229] A. Carrera Vazquez, C. Tornow, D. Ristè, S. Woerner, M. Takita, D.J. Egger, //dx.doi.org/10.1088/1367-2630/acf9c3.
Combining quantum processors with real-time classical communication, Nature [254] P. Gokhale, O. Angiuli, Y. Ding, K. Gui, T. Tomesh, M. Suchara, M. Martonosi,
(2024) 1–5, http://dx.doi.org/10.1038/s41586-024-08178-2. F.T. Chong, Minimizing state preparations in variational quantum eigensolver
[230] A. Lowe, M. Medvidović, A. Hayes, L.J. O’Riordan, T.R. Bromley, J.M. Arrazola, by partitioning into commuting families, 2019, http://dx.doi.org/10.48550/
N. Killoran, Fast quantum circuit cutting with randomized measurements, arXiv.1907.13623, arxiv preprint.
Quantum 7 (2023) 934, http://dx.doi.org/10.22331/q-2023-03-02-934. [255] M. Schuld, V. Bergholm, C. Gogolin, J. Izaac, N. Killoran, Evaluating analytic
[231] H.-Y. Huang, R. Kueng, J. Preskill, Predicting many properties of a quantum gradients on quantum hardware, Phys. Rev. A 99 (2019) 032331, http://dx.
system from very few measurements, Nat. Phys. 16 (2020) 1050–1057, http: doi.org/10.1103/PhysRevA.99.032331.
//dx.doi.org/10.1038/s41567-020-0932-7. [256] J. Stokes, J. Izaac, N. Killoran, G. Carleo, Quantum natural gradient, Quantum
[232] H. Harada, K. Wada, N. Yamamoto, Doubly optimal parallel wire cutting 4 (2020) 269, http://dx.doi.org/10.22331/q-2020-05-25-269.
without ancilla qubits, PRX Quantum 5 (4) (2024) 040308, http://dx.doi.org/ [257] J.D. Viqueira, D. Faílde, M.M. Juane, A. Gómez, D. Mera, Density matrix
10.1103/PRXQuantum.5.040308. emulation of quantum recurrent neural networks for multivariate time series
[233] E. Pednault, An alternative approach to optimal wire cutting without ancilla prediction, 2023, http://dx.doi.org/10.48550/arXiv.2310.20671, arXiv.
[258] D. Wierichs, J. Izaac, C. Wang, C.Y.-Y. Lin, General parameter-shift rules
qubits, 2023, http://dx.doi.org/10.48550/arXiv.2303.08287, arxiv preprint.
for quantum gradients, Quantum 6 (2022) 677, http://dx.doi.org/10.22331/q-
[234] C. Ufrecht, L.S. Herzog, D.D. Scherer, M. Periyasamy, S. Rietsch, A. Plinge, C.
2022-03-30-677.
Mutschler, Optimal joint cutting of two-qubit rotation gates, Phys. Rev. A 109
[259] D. Faílde, J.D. Viqueira, M. Mussa Juane, A. Gómez, Using differential evolution
(5) (2024) 052440, http://dx.doi.org/10.1103/PhysRevA.109.052440.
to avoid local minima in variational quantum algorithms, Sci. Rep. 13 (1)
[235] A.W. Harrow, A. Lowe, Optimal quantum circuit cuts with application to clus-
(2023) 16230, http://dx.doi.org/10.1038/s41598-023-43404-3.
tered Hamiltonian simulation, 2024, http://dx.doi.org/10.48550/arXiv.2403. [260] M.S. Alvarez-Alvarado, F.E. Alban-Chacón, E.A. Lamilla-Rubio, C.D. Rodríguez-
01018, arXiv. Gallegos, W. Velásquez, Three novel quantum-inspired swarm optimization
[236] L. Schmitt, C. Piveteau, D. Sutter, Cutting circuits with multiple two-qubit algorithms using different bounded potential fields, Sci. Rep. 11 (1) (2021)
unitaries, 2023, http://dx.doi.org/10.48550/arXiv.2312.11638, arxiv preprint. 11655, http://dx.doi.org/10.1038/s41598-021-90847-7.
[237] C. Ufrecht, M. Periyasamy, S. Rietsch, D.D. Scherer, A. Plinge, C. Mutschler, [261] E.R. Anschuetz, B.T. Kiani, Quantum variational algorithms are swamped with
Cutting multi-control quantum gates with ZX calculus, Quantum 7 (2023) 1147, traps, Nat. Commun. 13 (1) (2022) 7760, http://dx.doi.org/10.1038/s41467-
http://dx.doi.org/10.22331/q-2023-10-23-1147. 022-35364-5.
[238] J. Chen, E.M. Stoudenmire, S.R. White, Quantum Fourier transform has [262] Y. Du, Y. Qian, X. Wu, D. Tao, A distributed learning scheme for variational
small entanglement, PRX Quantum 4 (4) (2023) http://dx.doi.org/10.1103/ quantum algorithms, IEEE Trans. Quantum Eng. 3 (2022) http://dx.doi.org/10.
PRXQuantum.4.040318. 1109/TQE.2022.3175267.

27
D. Barral et al. Computer Science Review 57 (2025) 100747

[263] S.Y.C. Chen, S. Yoo, Federated quantum machine learning, Entropy 23 (4) [286] F. Chow, Intermediate representation, Queue 11 (2013) 30–37, http://dx.doi.
(2021) http://dx.doi.org/10.3390/e23040460. org/10.1145/2542661.2544374.
[264] S. Stein, N. Wiebe, Y. Ding, P. Bo, K. Kowalski, N. Baker, J. Ang, A. Li, [287] A. McCaskey, T. Nguyen, A MLIR dialect for quantum assembly languages,
EQC: Ensembled quantum computing for variational quantum algorithms, in: in: Int. Conf, on Quantum Computing and Engineering, QCE, IEEE, 2021, pp.
Proc. of the 49th Int. Symp. on Computer Architecture, ACM, 2022, pp. 59–71, 255–264, http://dx.doi.org/10.1109/QCE52317.2021.00043.
http://dx.doi.org/10.1145/3470496.3527434. [288] T. Lubinski, C. Granade, A. Anderson, A. Geller, M. Roetteler, A. Petrenko,
[265] P. Das, S.S. Tannu, P.J. Nair, M. Qureshi, A case for multi-programming B. Heim, Advancing hybrid quantum–classical computation with real-time
quantum computers, in: Proceedings of the 52nd Annual IEEE/ACM Int. Symp. execution, Front. Phys. 10 (2022) 940293, http://dx.doi.org/10.3389/FPHY.
on Microarchitecture, 2019, pp. 291–303, http://dx.doi.org/10.1145/3352460. 2022.940293.
3358287. [289] A. Peduri, S. Bhat, T. Grosser, QSSA: An SSA-based IR for quantum computing,
[266] A. Ash-Saki, M. Alam, S. Ghosh, Analysis of crosstalk in NISQ devices and in: Proc. of the 31st ACM SIGPLAN Int. Conf. on Compiler Construction,
security implications in multi-programming regime, in: Proc. of the ACM/IEEE ACM, New York, NY, USA, 2022, pp. 2–14, http://dx.doi.org/10.1145/3497776.
Int. Symp. on Low Power Electronics and Design, 2020, pp. 25–30, http: 3517772.
//dx.doi.org/10.1145/3370748.3406570. [290] D. Ittah, T. Häner, V. Kliuchnikov, T. Hoefler, QIRO: A static single assignment-
[267] L. Liu, X. Dou, QuCloud: A new qubit mapping mechanism for multi- based quantum program representation for optimization, ACM Trans. Quantum
programming quantum computing in cloud environment, in: IEEE Int. Symp. Comput. 3 (3) (2022) http://dx.doi.org/10.1145/3491247.
on High-Performance Computer Architecture, HPCA, IEEE, 2021, pp. 167–178, [291] K. Hietala, R. Rand, S.-H. Hung, X. Wu, M. Hicks, Verified optimization in a
http://dx.doi.org/10.1109/HPCA51647.2021.00024. quantum intermediate representation, 2019, http://dx.doi.org/10.48550/arXiv.
[268] A.A. Saki, S. Ghosh, Qubit sensing: A new attack model for multi- 1904.06319, arxiv preprint.
programming quantum computing, 2021, http://dx.doi.org/10.48550/arXiv. [292] X.-Z. Luo, J.-G. Liu, P. Zhang, L. Wang, Yao.jl: Extensible, efficient framework
2104.05899, arxiv preprint. for quantum algorithm design, Quantum 4 (2020) 341, http://dx.doi.org/10.
[269] S. Niu, A. Todri-Sanial, Multi-programming cross platform benchmarking for 22331/q-2020-10-11-341.
quantum computing hardware, 2022, http://dx.doi.org/10.48550/arXiv.2206. [293] S. Nishio, R. Wakizaka, InQuIR: Intermediate representation for interconnected
03144, arxiv preprint. quantum computers, 2023, http://dx.doi.org/10.48550/arXiv.2302.00267, arxiv
[270] S. Niu, A. Todri-Sanial, How parallel circuit execution can be useful for NISQ preprint.
computing? in: 2022 Design, Automation & Test in Europe Conference & Exhi- [294] A.W. Cross, L.S. Bishop, J.A. Smolin, J.M. Gambetta, Open quantum assembly
bition, DATE, 2022, pp. 1065–1070, http://dx.doi.org/10.23919/DATE54114. language, 2017, http://dx.doi.org/10.48550/arXiv.1707.03429, arxiv preprint.
2022.9774512. [295] N. Khammassi, G.G. Guerreschi, I. Ashraf, J.W. Hogaboam, C.G. Almudever, K.
[271] S. Niu, A. Todri-Sanial, Enabling multi-programming mechanism for quantum Bertels, cQASM v1.0: Towards a common quantum assembly language, 2018,
computing in the NISQ era, Quantum 7 (2023) 925, http://dx.doi.org/10. http://dx.doi.org/10.48550/arXiv.1805.09607, arxiv preprint.
22331/q-2023-02-16-925. [296] X. Fu, L. Riesebos, M.A. Rol, J. Van Straten, J. Van Someren, N. Khammassi, I.
[272] S. Niu, A. Todri-Sanial, Multi-programming mechanism on near-term quantum Ashraf, R.F. Vermeulen, V. Newsum, K.K. Loh, J.C. De Sterke, W.J. Vlothuizen,
computing, in: Quantum Computing: Circuits, Systems, Automation and Applica- R.N. Schouten, C.G. Almudever, L. Dicarlo, K. Bertels, EQASM: An executable
tions, Springer, 2023, pp. 19–54, http://dx.doi.org/10.1007/978-3-031-37966- quantum instruction set architecture, in: Proc. 25th IEEE Int. Symp. on High
6_2. Performance Computer Architecture, HPCA, IEEE, 2019, pp. 224–237, http:
[273] L. Liu, X. Dou, QuCloud+: A holistic qubit mapping scheme for single/multi- //dx.doi.org/10.1109/HPCA.2019.00040.
programming on 2D/3D NISQ quantum computers, ACM Trans. Archit. Code [297] K.M. Svore, A.V. Aho, A.W. Cross, I. Chuang, I.L. Markov, A layered software
Optim. 21 (1) (2024) 1–27, http://dx.doi.org/10.1145/3631525. architecture for quantum computing design tools, Computer 39 (1) (2006)
[274] J.K. Lee, O.T. Brown, M. Bull, M. Ruefenacht, J. Doerfert, M. Klemm, M. Schulz, 74–83, http://dx.doi.org/10.1109/MC.2006.4.
Quantum task offloading with the openmp API, 2023, http://dx.doi.org/10. [298] A. Dahlberg, B. van der Vecht, C.D. Donne, M. Skrzypczyk, I.t. Raa, W.
48550/arXiv.2311.03210, arxiv preprint. Kozlowski, S. Wehner, NetQASM - a low-level instruction set architecture
[275] The CUDA Quantum development team, CUDA Quantum, 2025, URL https: for hybrid quantum-classical programs in a quantum internet, Quantum Sci.
//github.com/NVIDIA/cuda-quantum. (Accessed 21 January 2025). Technol. 7 (3) (2022) 035023, http://dx.doi.org/10.1088/2058-9565/ac753f.
[276] A.J. Abhari, A. Faruque, M.J. Dousti, L. Svec, O. Catu, A. Chakrabati, C.-F. [299] M. Ying, Y. Feng, An algebraic language for distributed quantum computing,
Chiang, S. Vanderwilt, J. Black, F. Chong, M. Martonosi, M. Suchara, K. Brown, IEEE Trans. Comput. 58 (6) (2009) 728–743, http://dx.doi.org/10.1109/TC.
M. Pedram, T. Brun, Scaffold: Quantum Programming Language, Tech. Rep., 2009.13.
Princeton University, Dept. of Computer Science, 2012, URL https://www.cs. [300] X. Qiu, L. Chen, Quantum cost of dense coding and teleportation, Phys. Rev.
princeton.edu/techreports/2012/934.pdf. (Accessed 21 January 2025). A 105 (6) (2022) 062451, http://dx.doi.org/10.1103/PhysRevA.105.062451.
[277] K. Svore, A. Geller, M. Troyer, J. Azariah, C. Granade, B. Heim, V. Kliuchnikov, [301] C. Piveteau, D. Sutter, Circuit knitting with classical communication, IEEE
M. Mykhailova, A. Paz, M. Roetteler, Q#: Enabling scalable quantum computing Trans. Inform. Theory 70 (4) (2024) 2734–2745, http://dx.doi.org/10.1109/
and development with a high-level DSL, in: Proc. of the Real World Domain TIT.2023.3310797.
Specific Languages Workshop, ACM, 2018, pp. 1–10, http://dx.doi.org/10. [302] D.W. Wall, Global register allocation at link time, ACM SIGPLAN Not. 21 (1986)
1145/3183895.3183901. 264–275, http://dx.doi.org/10.1145/13310.13338.
[278] J. Guo, H. Lou, J. Yu, R. Li, W. Fang, J. Liu, P. Long, S. Ying, M. Ying, isQ: [303] T. Ito, N. Kakimura, N. Kamiyama, Y. Kobayashi, Y. Okamoto, Algorithmic
An integrated software stack for quantum programming, IEEE Trans. Quantum theory of qubit routing, in: Algorithms and Data Structures, Springer Nature,
Eng. 4 (2023) http://dx.doi.org/10.1109/TQE.2023.3275868. 2023, pp. 533–546, http://dx.doi.org/10.1007/978-3-031-38906-1_35.
[279] S. Liu, X. Wang, L. Zhou, J. Guan, Y. Li, Y. He, R. Duan, M. Ying, Q|SI⟩: A [304] A. Paler, On the influence of initial qubit placement during NISQ circuit
quantum programming environment, in: Symposium on Real-Time and Hybrid compilation, in: Quantum Technology and Optimization Problems, Springer,
Systems: Essays Dedicated to Professor Chaochen Zhou on the Occasion of 2019, pp. 207–217, http://dx.doi.org/10.1007/978-3-030-14082-3_18.
His 80th Birthday, Springer International Publishing, 2018, pp. 133–164, http: [305] M. Bandic, C.G. Almudever, S. Feld, Interaction graph-based characterization
//dx.doi.org/10.1007/978-3-030-01461-2_8, Chapter 8. of quantum benchmarks for improving quantum circuit mapping techniques,
[280] Qiskit contributors, Qiskit: An open-source framework for quantum computing, Quantum Mach. Intell. 5 (2023) 1–30, http://dx.doi.org/10.1007/S42484-023-
2023, http://dx.doi.org/10.5281/zenodo.2573505. 00124-1.
[281] Cirq Developers, Cirq, 2023, http://dx.doi.org/10.5281/zenodo.4062499. [306] A. Cowtan, S. Dilkes, R. Duncan, A. Krajenbrink, W. Simmons, S. Sivarajah,
[282] Y. Suzuki, Y. Kawase, Y. Masumura, Y. Hiraga, M. Nakadai, J. Chen, K.M. On the qubit routing problem, in: 14th Conf. on the Theory of Quantum
Nakanishi, K. Mitarai, R. Imai, S. Tamiya, T. Yamamoto, T. Yan, T. Kawakubo, Computation, Communication and Cryptography, TQC, Vol. 135, 2019, pp.
Y.O. Nakagawa, Y. Ibe, Y. Zhang, H. Yamashita, H. Yoshimura, A. Hayashi, 5:1–5:32, http://dx.doi.org/10.4230/LIPIcs.TQC.2019.5.
K. Fujii, Qulacs: A fast and versatile quantum circuit simulator for research [307] H. Barnes, A Survey of Qubit Routing Algorithms (Ph.D. thesis), Rochester
purpose, Quantum 5 (2021) http://dx.doi.org/10.22331/Q-2021-10-06-559. Institute of Technology, 2023, URL https://repository.rit.edu/theses/11428.
[283] T. Häner, D.S. Steiger, T. Hoefler, M. Troyer, Distributed quantum computing [308] Y. Mao, Y. Liu, Y. Yang, Qubit allocation for distributed quantum computing,
with QMPI, in: Int. Conf. for High Perf. Computing, Networking, Storage and in: Conf. on Computer Communications, IEEE, 2023, pp. 1–10, http://dx.doi.
Analysis, IEEE, 2021, http://dx.doi.org/10.1145/3458817.3476172. org/10.1109/INFOCOM53939.2023.10228915.
[284] Y. Shi, T. Nguyen, S. Stein, T. Stavenger, M. Warner, M. Roetteler, T. Hoefler, [309] Y. Mao, Y. Liu, Y. Yang, Probability-aware qubit-to-processor mapping in
A. Li, A reference implementation for a quantum message passing interface, in: distributed quantum computing, in: 1st Workshop on Quantum Networks and
Proc. of the SC’23 Workshops, ACM, 2023, pp. 1420–1425, http://dx.doi.org/ Distributed Quantum Computing, ACM, 2023, pp. 51–56, http://dx.doi.org/10.
10.1145/3624062.3624212. 1145/3610251.3610554.
[285] R. Wakizaka, Towards reliable distributed quantum computing on quantum [310] M. Nakai, Qubit Allocation For Distributed Quantum Computing (Bachelor’s
interconnects, ACM Int. Conf. Proc. Ser. (2023) 114–116, http://dx.doi.org/ thesis), Keio University, 2021, URL https://aqua.sfc.wide.ad.jp/publications/
10.1145/3594671.3594691. dave_bthesis.pdf.

28
D. Barral et al. Computer Science Review 57 (2025) 100747

[311] Z. Chen, X. Chen, Y. Jiang, X. Cheng, Z. Guan, Routing strategy for distributed [329] S. Sivarajah, S. Dilkes, A. Cowtan, W. Simmons, A. Edgington, R. Duncan,
quantum circuit based on optimized gate transmission direction, Internat. J. T|ket⟩: A retargetable compiler for NISQ devices, Quantum Sci. Technol. 6 (1)
Theoret. Phys. 62 (12) (2023) 255, http://dx.doi.org/10.1007/s10773-023- (2020) 014003, http://dx.doi.org/10.1088/2058-9565/ab8e92.
05489-4. [330] J.C. Boschero, N.M.P. Neumann, W. van der Schoot, T. Sijpesteijn, R. Wezeman,
[312] S. Rodrigo, D. Spanò, M. Bandic, S. Abadal, H. Van Someren, A. Ovide, S. Feld, Distributed quantum computing: Applications and challenges, 2024, http://dx.
C.G. Almudéver, E. Alarcón, Characterizing the spatio-temporal qubit traffic of doi.org/10.48550/arXiv.2410.00609.
a quantum intranet aiming at modular quantum computer architectures, in: 9th [331] C. Gidney, M. Ekerå, How to factor 2048 bit RSA integers in 8 hours using 20
ACM Int. Conf. on Nanoscale Computing and Communication, 2022, pp. 1–7, million noisy qubits, Quantum 5 (2021) http://dx.doi.org/10.22331/q-2021-04-
http://dx.doi.org/10.1145/3558583.3558846. 15-433.
[313] S.B. Rached, I.L. Agudo, S. Rodrigo, M. Bandic, S. Feld, H. van Someren, [332] M. Ekerå, J. Håstad, Quantum algorithms for computing short discrete loga-
E. Alarcón, C.G. Almudéver, S. Abadal, Characterizing the inter-core qubit rithms and factoring RSA integers, in: T. Lange, T. Takagi (Eds.), Post-Quantum
traffic in large-scale quantum modular architectures, 2023, http://dx.doi.org/ Cryptography, in: LNCS, no. 10346, Springer, 2017, pp. 347–363, http://dx.doi.
10.48550/arXiv.2310.01921, arxiv preprint. org/10.1007/978-3-319-59879-6_20.
[314] N. Khammassi, I. Ashraf, J. Someren, R. Nane, A. Krol, M.A. Rol, L. Lao, K. [333] L. Xiao, D. Qiu, L. Luo, P. Mateus, Distributed quantum-classical hybrid shor’s
Bertels, C.G. Almudever, OpenQL: A portable quantum programming framework algorithm, 2023, http://dx.doi.org/10.48550/arXiv.2304.12100, arxiv preprint.
for quantum accelerators, ACM J. Emerg. Technol. Comput. Syst. 18 (1) (2021) [334] N.M. Neumann, R. van Houte, T. Attema, Imperfect distributed quantum phase
1–24, http://dx.doi.org/10.1145/3474222. estimation, in: ICCS, in: LNCS, vol. 12142, Springer, 2020, pp. 605–615,
[315] L. Lao, H. Van Someren, I. Ashraf, C.G. Almudever, Timing and resource- http://dx.doi.org/10.1007/978-3-030-50433-5_46.
aware mapping of quantum circuits to superconducting processors, IEEE Trans. [335] R. Van Meter, W.J. Munro, K. Nemoto, K.M. Itoh, Arithmetic on a distributed-
Comput.- Aided Des. Integr. Circuits Syst. 41 (2) (2021) 359–371, http://dx. memory quantum multicomputer, ACM J. Emerg. Technol. Comput. Syst. 3 (4)
doi.org/10.1109/TCAD.2021.3057583. (2008) http://dx.doi.org/10.1145/1324177.1324179.
[316] A. Fagan, R. Duncan, Optimising clifford circuits with quantomatic, Electron. [336] J. Tan, L. Xiao, D. Qiu, L. Luo, P. Mateus, Distributed quantum algorithm
Proc. Theor. Comput. Sci. 287 (2019) 85–105, http://dx.doi.org/10.4204/eptcs. for Simon’s problem, Phys. Rev. A 106 (3) (2022) http://dx.doi.org/10.1103/
287.5. PhysRevA.106.032417.
[317] M. Xu, Z. Li, O. Padon, S. Lin, J. Pointing, A. Hirth, H. Ma, J. Palsberg, A. Aiken, [337] H. Li, D. Qiu, L. Luo, Distributed exact quantum algorithms for Deutsch–Jozsa
U.A. Acar, Z. Jia, Quartz: superoptimization of quantum circuits, in: Proc. of problem, 2023, http://dx.doi.org/10.48550/arXiv.2303.10663, arxiv preprint.
the Conf. on Programming Language Design and Implementation, PLDI, ACM, [338] X. Zhou, D. Qiu, L. Luo, Distributed Bernstein–Vazirani algorithm, Phys. A 629
2022, pp. 625–640, http://dx.doi.org/10.1145/3519939.3523433. (2023) 129209, http://dx.doi.org/10.1016/j.physa.2023.129209.
[318] S. Haider, S.A.R. Kazmi, An extended quantum process algebra (eQPAlg) [339] X. Zhou, D. Qiu, L. Luo, Distributed exact Grover’s algorithm, Front. Phys. 18
approach for distributed quantum systems, 2020, http://dx.doi.org/10.48550/ (5) (2023) 1–25, http://dx.doi.org/10.1007/s11467-023-1327-x.
arXiv.2001.04249, arxiv preprint. [340] J. Avron, O. Casper, I. Rozen, Quantum advantage and noise reduction in
[319] M. Lalire, P. Jorrand, A process algebraic approach to concurrent and dis- distributed quantum computing, Phys. Rev. A 104 (2021) http://dx.doi.org/
tributed quantum computation: Operational semantics, 2004, http://dx.doi.org/ 10.1103/PhysRevA.104.052404.
10.48550/arXiv.quant-ph/0407005, arXiv. [341] K. Fujii, K. Mizuta, H. Ueda, K. Mitarai, W. Mizukami, Y.O. Nakagawa, Deep
[320] Y. Feng, S. Li, M. Ying, Verification of distributed quantum programs, ACM variational quantum eigensolver: A divide-and-conquer method for solving a
Trans. Comput. Log. 23 (3) (2022) http://dx.doi.org/10.1145/3517145. larger problem with smaller size quantum computers, PRX Quantum 3 (2022)
[321] C.A.R. Hoare, An axiomatic basis for computer programming, Commun. ACM http://dx.doi.org/10.1103/PRXQuantum.3.010346.
12 (1969) 576–580, http://dx.doi.org/10.1145/363235.363259. [342] Z. Zhou, Y. Du, X. Tian, D. Tao, QAOA-in-QAOA: Solving large-scale MaxCut
[322] Y. Wang, Verification of distributed quantum protocols, 2021, http://dx.doi. problems on small quantum machines, Phys. Rev. Appl. 19 (2) (2023) 024027,
org/10.48550/arXiv.2110.11416, arxiv preprint. http://dx.doi.org/10.1103/PhysRevApplied.19.024027.
[323] C.H. Bennett, G. Brassard, Quantum cryptography: Public key distribution and [343] S.C. Marshall, C. Gyurik, V. Dunjko, High dimensional quantum machine
coin tossing, Theoret. Comput. Sci. 560 (2014) 7–11, http://dx.doi.org/10. learning with small quantum computers, Quantum 7 (2022) http://dx.doi.org/
1016/j.tcs.2014.05.025. 10.22331/q-2023-08-09-1078.
[324] D. Ferrari, S. Nasturzio, M. Amoretti, Poster: A software tool for [344] K. Zhang, P. Rao, K. Yu, H. Lim, V. Korepin, Implementation of efficient
mapping and executing distributed quantum computations on a net- quantum search algorithms on NISQ computers, Quantum Inf. Process. 20
work simulator, in: QCrypt, 2021, URL https://2021.qcrypt.net/posters/ (2021) 233, http://dx.doi.org/10.1007/s11128-021-03165-2.
QCrypt2021Poster185Ferrari.pdf. (Accessed 21 January 2025). [345] G. Park, K. Zhang, K. Yu, V. Korepin, Quantum multi-programming for Grover’s
[325] D. Ferrari, S. Carretta, M. Amoretti, A modular quantum compilation framework search, Quantum Inf. Process. 22 (1) (2023) http://dx.doi.org/10.1007/s11128-
for distributed quantum computing, IEEE Trans. Quantum Eng. 4 (2023) 1–13, 022-03793-2.
http://dx.doi.org/10.1109/TQE.2023.3303935. [346] L.K. Grover, Trade-offs in the quantum search algorithm, Phys. Rev. A 66 (2002)
[326] D. Cuomo, Architectures and Circuits for Distributed Quantum Computing http://dx.doi.org/10.1103/PhysRevA.66.052314.
(Ph.D. thesis), Univ. of Naples Federico II, Univ. of Camerino, National Research [347] K. Li, D. Qiu, L. Li, S. Zheng, Z. Rong, Application of distributed semi-quantum
Council of Italy, 2023, URL http://www.fedoatd.unina.it/961/1/THESIS_DQC. computing model in phase estimation, Inform. Process. Lett. 120 (2017) 23–29,
pdf. http://dx.doi.org/10.1016/j.ipl.2016.12.002.
[327] Y. Ohkura, T. Satoh, R. Van Meter, Simultaneous execution of quantum circuits [348] T. Tanaka, Y. Suzuki, S. Uno, R. Raymond, T. Onodera, N. Yamamoto,
on current and near-future NISQ systems, IEEE Trans. Quantum Eng. 3 (2022) Amplitude estimation via maximum likelihood on noisy quantum computer,
1–10, http://dx.doi.org/10.1109/TQE.2022.3164716. Quantum Inf. Process. 20 (2021) 1–29, http://dx.doi.org/10.1007/s11128-021-
[328] T. Tomesh, Z.H. Saleem, M.A. Perlin, P. Gokhale, M. Suchara, M. Martonosi, 03215-9.
Divide and conquer for combinatorial optimization and distributed quantum [349] S. DiAdamo, M. Ghibaudi, J. Cruise, Distributed quantum computing and
computation, in: Int. Conf. on Quantum Computing and Engineering, QCE, Vol. network control for accelerated VQE, IEEE Trans. Quantum Eng. 2 (2021) 1–21,
1, IEEE, 2023, pp. 1–12, http://dx.doi.org/10.1109/QCE57702.2023.00009. http://dx.doi.org/10.1109/TQE.2021.3057908.

29