0% found this document useful (0 votes)
9 views8 pages

A Comparative Study of Blockchain Development Platforms ICCIP

The paper presents a comparative study of 12 popular blockchain development platforms, assessing their characteristics, capabilities, and limitations using a scoring approach. Key metrics evaluated include popularity, adoption, and maturity, which consider factors such as consensus mechanisms, programming languages, scalability, and security. The study aims to facilitate the selection of the most suitable platform for developers based on these comprehensive evaluations.

Uploaded by

Raja Rajagopal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
9 views8 pages

A Comparative Study of Blockchain Development Platforms ICCIP

The paper presents a comparative study of 12 popular blockchain development platforms, assessing their characteristics, capabilities, and limitations using a scoring approach. Key metrics evaluated include popularity, adoption, and maturity, which consider factors such as consensus mechanisms, programming languages, scalability, and security. The study aims to facilitate the selection of the most suitable platform for developers based on these comprehensive evaluations.

Uploaded by

Raja Rajagopal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/380059420

A Comparative Study of Blockchain Development Platforms

Conference Paper · April 2024


DOI: 10.1145/3638884.3638971

CITATIONS READS
0 133

4 authors:

Kenza Riahi Amine Brahmia


Université de Haute-Alsace & CESI Lineact CESI
8 PUBLICATIONS 17 CITATIONS 42 PUBLICATIONS 153 CITATIONS

SEE PROFILE SEE PROFILE

Abdelhafid Abouaissa Lhassane Idoumghar


University of Upper Alsace University of Upper Alsace
100 PUBLICATIONS 652 CITATIONS 217 PUBLICATIONS 6,436 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Kenza Riahi on 09 May 2024.

The user has requested enhancement of the downloaded file.


A Comparative Study of Blockchain Development Platforms
Kenza RIAHI Mohamed-el-Amine Brahmia
[email protected] [email protected]
CESI LINEACT UR 7527 CESI LINEACT UR 7527
Strasbourg, France Strasbourg, France
[email protected]
IRIMAS UR 7499, University of Haute Alsace
Mulhouse, France

Abdelhafid Abouaissa Lhassane Idoumghar


[email protected] [email protected]
IRIMAS UR 7499, University of Haute Alsace IRIMAS UR 7499, University of Haute Alsace
Mulhouse, France Mulhouse, France
ABSTRACT Developing a blockchain-based solution first requires under-
Blockchain technology is nowadays applied to several domains, standing the domain and the solution requirements but also regula-
essentially finance, healthcare, supply chain, and digital identity tions and standards in order to select the most adapted platform. As
management. For enterprises to adopt a blockchain solution, many early as the concept of blockchain was proposed in 2009 by Satoshi
existing platforms facilitate the development process rather than Nakamoto, many blockchain development platforms emerged, try-
implementing the system from scratch. Each platform has its own ing to meet enterprises’ and individuals’ needs. Each one is adapted
characteristics, capabilities, and limitations. Thus, several factors to a specific use case and has particular characteristics, advantages,
should be considered, including the consensus mechanism, the and drawbacks.
programming language, scalability, and security, as well as the In this paper, we conduct a comparative study of the 12 most
popularity and maturity of the platform. Using a scoring approach, used blockchain platforms. We propose metrics and scores to assess
we provide a comparative analysis of the 12 most commonly used and evaluate their popularity, adoption, and maturity. The maturity
platforms, which facilitates comparison and selection of the most score evaluates the platform scalability, security, maintenance, doc-
suitable platform for a developer’s application. umentation, and future improvements proposed by the developers.
These three scores have been designed to cover the most impor-
KEYWORDS tant aspects to consider while selecting a blockchain development
platform.
Blockchain platform, consensus protocol, scalability, popularity,
The rest of the paper is structured as follows: section 2 presents
maturity.
some literature works that surveyed blockchain platforms, section 3
ACM Reference Format: defines some concepts related to blockchain platforms such as con-
Kenza RIAHI, Mohamed-el-Amine Brahmia, Abdelhafid Abouaissa, and Lhas- sensus protocols and smart contracts. We also present the metrics
sane Idoumghar. 2024. A Comparative Study of Blockchain Development used in our study. In section 4, we present the 12 platforms with
Platforms. In . ICCIP, Lingshui, Hainan, China, 7 pages. https://doi.org/ their specific characteristics. Section 5 establishes a detailed com-
parison on the basis of the metrics we defined. We conclude the
1 INTRODUCTION paper in section 6.
Blockchain technology has many advantages of being decentralized,
immutable, tamper-proof, and transparent thanks to the crypto-
graphic mechanisms it uses and the P2P communication strategy. 2 RELATED WORK
Initially, it was used in the financial sector, but was later applied to
supply chain, healthcare, and energy sectors. Consequently, many Many previous works proposed comprehensive surveys on Blockchain
enterprises are converting their solutions to blockchain-based plat- platforms. We note that we used some information from them. The
forms in order to enhance security, transparency, and traceability. authors in [14] establish a complete comparison between the 23
most popular platforms on the basis of many parameters. The arti-
cle proposes 7 tables each one dealing with a specific aspect. Many
Permission to make digital or hard copies of all or part of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed scores were also proposed which facilitate the comparison and
for profit or commercial advantage and that copies bear this notice and the full citation understanding of each platform concept. However, some of the
on the first page. Copyrights for components of this work owned by others than the platforms stated in the survey are not very popular and their use is
author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or
republish, to post on servers or to redistribute to lists, requires prior specific permission very limited. Saraf et al. [26] only compare Hyperledger platforms
and/or a fee. Request permissions from [email protected]. (Burrow, Fabric, Indy, Iroha, and Sawtooth), Ethereum, and Corda.
ICCCP’23, December 2023, Lingshui, Hainan, China Although the platforms are described in detail, their popularity,
© 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.
ACM ISBN adoption, documentation quality, maintenance, and roadmap are
https://doi.org/ not assessed.
ICCCP’23, December 2023, Lingshui, Hainan, China Riahi et al.

The authors of [13] present a brief survey of five blockchain • Proof-of-Stake (PoS): As its name suggests, it is the amount
development platforms (Hyperledger, VeChain, Ripple, R3Corda, of stake (or coins) at possession that defines the most likely
and Ethereum). Although the platforms are well presented, the peers to add a new block.
survey does not provide an accurate comparison between them. • Proof-of-Elapsed-Time (PoET): This protocol consists of each
Another work [18] compares the 5 most used platforms: Ethereum, peer generating a random number that represents the wait-
Hyperledger Fabric, MultiChain, Quorum, and Corda. However, a ing time before being able to validate a block[12]. It is, as a
comparison regarding popularity, adoption, and maturity is not result, fair since all the nodes can validate a block, it is also
stated. efficient because peers do not compete to reach an agree-
ment.
3 PRELIMINARIES • Practical Byzantine Fault Tolerance (PBFT): This protocol tol-
In this section, we introduce the essential concepts of blockchain erates up to 1/3 byzantine nodes thanks to its three-step pro-
technology and we present the metrics we chose to assess the cess: A transaction has to pass through pre-prepare, prepare,
blockchain platforms. and commit phases to be validated. It works upon partially
synchronous systems and has an interesting throughput in
comparison to other protocols. However, its main disadvan-
3.1 Blockchain tage is the heavy network load because of the amount of
Blockchain technology is a way of storing and managing data and communication required[25].
transactions within a distributed ledger: Each entity that takes part
in the system maintains a copy of the blockchain, a copy that is 3.4 Smart Contracts
similar for all the peers.
Data and transactions are packed into chained blocks in such a Smart contracts are self-executing computer programs or virtual
way that each block stores the hash of the previous block, and cryp- third parties[18]. Concretely, they are small programs stored in
tography ensures the transparency and immutability of the system. the blockchain itself -which makes them immutable- and executed
Asymmetric encryption, hash techniques, and digital signatures once certain conditions are fulfilled. Thanks to them, there is no
are used to guarantee authentication, integrity, confidentiality, and need for physical third-party intermediaries which reduces delays
non-repudiation. and costs and enhances security.
Packing data into blocks uses the Merkle tree which allows the
definition of one hash for multiple transactions, each two adjacent 3.5 Comparision metrics
transactions are given a hash, and so forth according to a binary In what follows, we present the metrics we estimate are essential
tree. The ordering of the transactions and validating them relies on for selecting an adapted blockchain development platform.
a consensus protocol which allows the majority of participants to
• Popularity: This metric shows community interest in using
agree on a common value or state.
the platform. [14] establishes a popularity score based on
GitHub forks (f), GitHub stars (s), StackExchange questions
3.2 Blockchain types (q), and Google Scholar papers. [23] adds to these metrics the
The blockchain type defines how users can access the system. It number of contributors in Github (c), the number of commits
depends on the type of data or transactions and on the application (m), the number of followers on Twitter (l), and the number
domain. There are two main blockchain types: public (permission- of tweets in the official account (t). We will use 7 of these
less) and private (permissioned). In public blockchains, any node metrics (the number of Google Scholar papers will be used
can join the system, read the transactions, and participate in the later) to propose a score calculated as their normalized sum:
mining process. The most well-known public blockchains are Bit-
coin and Ethereum. Whereas in private blockchains, access rights
𝑓𝑖 𝑚𝑖 𝑐𝑖 𝑠𝑖 𝑞𝑖 𝑙𝑖 𝑡𝑖
are managed by one or many entities. Ripple and Hyperledger are 𝑓𝑚𝑎𝑥 𝑚𝑚𝑎𝑥 𝑐𝑚𝑎𝑥 𝑠𝑚𝑎𝑥 𝑞𝑚𝑎𝑥 𝑙𝑚𝑎𝑥 𝑡𝑚𝑎𝑥
examples of private blockchains. 𝑃𝑜𝑝𝑢𝑙𝑎𝑟𝑖𝑡𝑦𝑖 = (1)
7

3.3 Consensus Protocols • Adoption: This is an interesting metric proposed in [23]. It


gives an idea of the platform’s uses in the industrial and aca-
A transaction should be validated by the majority of the nodes in demic domains. The number of recent industrial projects us-
the system through a process called the consensus protocol in order ing each platform was taken from the 2019 Forbes Blockchain
to be added to the blockchain. In what follows, we will define the 50 report. In this paper, we will rely on their recent 2022
most used algorithms: report[4] to define the number of projects that used each
• Proof-of-Work (PoW): When blockchain was first proposed platform (p).
to solve the centrality problem of financial transactions man- Then, using the number of Google Scholar papers (s), we
agement, it used the PoW consensus protocol that consists can define the interest in the platform within the academic
of competing to solve a cryptographic puzzle by finding community. For this purpose, we entered the platform name
the good nonce. It is a brute force algorithm, only solvable followed by "blockchain platform" so as not to be confused
by iteratively trying different nonces. This makes it very with related topics. The adoption score will be calculated as
inefficient in terms of energy consumption [11]. the normalized sum of these two metrics:
A Comparative Study of Blockchain Development Platforms ICCCP’23, December 2023, Lingshui, Hainan, China

contracts which can be programmed using the object-oriented lan-


𝑝𝑖 𝑠𝑖 guage Solidity[22]. In order to execute them, the sender has to
𝐴𝑑𝑜𝑝𝑡𝑖𝑜𝑛𝑖 =
𝑝𝑚𝑎𝑥 𝑠𝑚𝑎𝑥
(2) pay transaction fees in 𝐸𝑡ℎ𝑒𝑟 cryptocurrency. The more amount
2 of 𝐸𝑡ℎ𝑒𝑟 is proposed by the transaction proposer the more miners
• Maturity: This metric indicates the development level of the are interested in validating it. Ethereum is composed of accounts,
project. A mature platform should be easy to handle either and state transitions represent the transfer of value or information
for developers or for users. Tests should be complete so that from one account to another. It first used PoW as a consensus pro-
bugs will not be encountered. In order to measure it, we tocol, then moved to PoS which reduced energy consumption by
utilize the following indicators partially based on the Capa- 99.95%[8]. In fact, it is a public blockchain that can be accessed
bility Maturity Model Integration (CMMI) for IT maturity by any entity without the need for registration. However, it can
proposed by ISACA[2] used to evaluate software systems. also be configured as a private blockchain. It was also the first
For each indicator, we will give a score from 1 to five and to propose a virtual machine blockchain called Ethereum Virtual
calculate the overall maturity score as their normalized sum. Machine (EVM) which makes its validation delay very interesting:
– Documentation: A criterion that should be taken into con- about 15 transactions per second (TPS). A block requires 12 to 15
sideration is the quality and amount of documentation seconds to be added to the ledger [20]. Ethereum is widely used in
provided which facilitates the task for developers and en- the financial domain[27] but also for document management[21].
courages them to choose that platform.
– Security and privacy: As transactions are broadcasted over
4.2 Hyperledger
the network, validated by the majority of parties, and
stored at each node level, the network and storage security Hyperledger (Umbrella project) is a multi-project open-source plat-
should be measured as well as the privacy within the form for permissioned blockchain applications created by the Linux
system, including access and user management. Foundation in December 2015. It was programmed with Golang and
– Scalability and performance: The platform scalability re- Java programming languages. It supports smart contracts which
garding both network load and storage should be consid- are written in chain code. The membership service provider (MSP)
ered. allows parties to join the network after their access rights have
– Maintenance: A platform that is continuously improved, been defined. Hyperledger has been especially used in Business-to-
upgraded, and optimized would be more interesting. Business (B2B) projects in supply chain, banking, healthcare, and
– Roadmap: A platform having a clear roadmap would be governmental applications.
better especially when the developer aims at deploying Hyperledger proposes many solutions for developing permis-
his projects to a production level. sioned blockchain platforms [13]:
• Network type: Some platforms are dedicated to public blockchains,
whereas others only support private blockchains. This factor • Fabric: Hyperledger Fabric allows the development of per-
should be taken into consideration when choosing a plat- missioned blockchains using a modular architecture. It sup-
form. ports multiple consensus protocols which can be adjusted
• Consensus protocol: Each platform uses a specific protocol depending on the developer’s needs (PBFT, Raft, Kafka and
for reaching consensus among the participants. Developers Solo). With Fabric, it is possible to create channels to restrict
should rely on it since each protocol operates in a different access rights only to some participants regarding confiden-
way and is conceived for a defined use case. tial transactions. Fabric’s participants can be clients who
• Smart Contracts Development Language: All the platforms propose transactions to be validated, peers who maintain
we present in this paper handle smart contracts. The sup- the ledger by performing the consensus protocol, or order-
ported development languages, however, are different from ing services responsible for ordering transactions within the
one platform to another and it would be more suitable some- blocks [23]. The authors in [18] consider that Hyperledger
times for developers to use a language they already master. Fabric remains the most interesting platform thanks to its
• Throughput: It can be represented by the number of vali- higher throughput and its flexibility regarding consensus
dated transactions per second. Some domains require more protocols but also concerning programming languages for
interesting throughput especially when handling real-time developing smart contracts with Java, Go, JavaScript and
transactions. TypeScript.
• Sawtooth: Sawtooth supports both permissioned and permis-
4 BLOCKCHAIN PLATFORMS sionless blockchain networks. It uses Proof-of-Elapsed-Time
(PoET) as a consensus protocol but also supports other proto-
In this section, we present the 12 most used blockchain development cols such as PBFT. Smart contracts can be written in almost
platforms with their different characteristics. any language of our choice (Go, Python, Java, JavaScript,
etc.).
4.1 Ethereum • Burrow: This platform allows the deployment of Ethereum
Ethereum is a transaction-based state machine that begins with a blockchains privately. It uses the Byzantine fault-tolerant
genesis state and incrementally executes a transaction to modify it Tendermint consensus protocol[24]. Its GitHub repository
into a final state[26]. It was the first project that introduced smart has been archived by Hyperledger on May 2022.
ICCCP’23, December 2023, Lingshui, Hainan, China Riahi et al.

• Indy: It uses digital identity to manage access rights and node that stores a copy of the whole ledger, or a light node that
employs the RBFT (Redundant BFT) consensus protocol[10]. performs light client verification.
• Iroha: Iroha is a project hosted by the Linux Foundation un-
der the Hyperledger project developed in C++ programming 4.8 Zilliqa
language[15]. The used consensus protocol is BFT. Iroha is Zilliqa is a blockchain platform dedicated to permissionless projects.
especially used when there is a need for the development of It is scalable because it uses sharding which divides network nodes
mobile applications. into sub-networks capable of processing transactions in parallel
[28].
4.3 R3 Corda
Corda is a distributed ledger platform dedicated to financial appli- 5 COMPARISION OF BLOCKCHAIN
cations but also used in digital assets, insurance, supply chain, PLATFORMS
and healthcare[19]. The official website describes it as "both a This part serves as a comparison and a guide to effectively select
blockchain and not a blockchain" because transactions are crypto- the most adapted platform to the developer’s needs.
graphically linked but not periodically packed into blocks. Smart We first evaluate the platforms’ popularity using the score de-
contracts can be developed with Kotlin or Java programming lan- fined in section 3 which uses different parameters: number of
guages. It is a permissioned platform where the 𝐷𝑜𝑜𝑟𝑚𝑎𝑛 requests GitHub forks, number of commits, contributors, GitHub stars, Stack-
authentication for nodes to access the network. Generally, it is Exchange questions, Twitter followers, and number of tweets. Their
used for B2B (business-to-business) projects, especially in finance, values on table 1 correspond to data available on September 2023.
supply chain, and healthcare. It is supported by Bank of America, The table shows the difference in popularity between the 12
HSBC, Intel and Microsoft. platforms. Ethereum is by far the most popular platform based
on our score calculation (with a score of 0.90). It has the highest
4.4 Quorum values for almost all the criteria. The second most popular platform
Quorum is a blockchain platform built on Ethereum [18] by fi- is Hyperledger Fabric (0.48) followed by Hyperledger Indy (0.31),
nancial services company JPMorgan. Unlike Ethereum, Quorum Quorum (0.28), Corda (0.27), and Stellar (0.22).
transactions can be made private and only be accessible to some The second aspect we evaluate is the platform adoption which
predefined users in the transaction itself. It also offers additional is based on its use in both the academic and the industrial domains.
possibilities regarding consensus protocols: Raft and IBFT (Istanbul The adoption score we defined in section 3 is evaluated for each
BFT) are supported as well. platform and displayed in table 2. It shows that Ethereum is the
most used platform with a score of 1, followed by Hyperledger
4.5 Stellar Fabric with a score of 0.49. The other platforms are less used in
Stellar is an open-source platform for payment proposed in 2014. industrial projects and are not cited as much as the top 2 platforms
It supports all types of assets (Dollar, Euro, Bitcoin, etc.) thanks to with scores less than 0.12).
anchors. It facilitates money transfer between countries and avoids Regarding maturity, the radar chart of figure 1 represents the
third parties which reduces costs and delays. Stellar can reduce up score levels for each aspect seen in section 3.
to 10% of the transaction fees. There is no need to develop smart
contracts in Stellar since everything is already developed in the
Ethereum
platform. The used consensus protocol is FBA which consists of
Hyperledger Fabric
a candidate node nomination followed by a federated voting [17].
Hyperledger Sawtooth
Stellar has an interesting latency and throughput: the confirmation
Hyperledger Burrow
of a transaction takes less than 5 seconds and the network can
Hyperledger Indy
process thousands of transactions per second [14].
Hyperledger Iroha

4.6 MultiChain R3 Corda

Quorum
Multichain[16] is an open-source blockchain protocol built on Bit-
Stellar
coin by adding additional possibilities regarding access rights man-
agement. In MultiChain, two or more nodes can create their own MultiChain

private chain which can only be accessed by authorized users. In Tendermint

each chain, only some nodes have the right to mine blocks with Zilliqa

the PoW consensus algorithm. In MultiChain 2.0, smart contracts


are called "Smart Filters" and can be developed with JavaScript.

4.7 Tendermint
Tendermint is a part of the Cosmos project which proposes an Figure 1: Platforms maturity chart
application-based blockchain that separates the application layer
from the consensus and networking layers[9]. A Tendermint node In what follows, we analyze each one of the aspects that served
can be of three types: a validator that validates transactions, a full to evaluate the platforms’ maturity, which consists of the quality
A Comparative Study of Blockchain Development Platforms ICCCP’23, December 2023, Lingshui, Hainan, China

Table 1: Popularity score calculation

Platform Forks Commits Contributors GitHub Stack ques- Followers Tweets Popularity
stars tions score

Ethereum 18400 14474 935 43500 42404 3100000 3561 0.90


Hyperledger 8600 14433 342 14800 6888 79200 12700 0.48
Fabric
Hyperledger 784 8255 78 1400 294 79200 12700 0.25
Sawtooth
Hyperledger 363 2742 38 1000 44 79200 12700 0.19
Burrow
Hyperledger 746 13406 147 668 155 79200 12700 0.31
Indy
Hyperledger 267 1181 51 389 57 79200 12700 0.17
Iroha
R3 Corda 1100 10140 194 4000 5396 57200 8345 0.27
Quorum 1300 14360 499 4500 374 13200 3543 0.28
Stellar 1100 8838 89 3100 127 753000 6075 0.22
MultiChain 271 1454 4 533 147 214100 2955 0.06
Tendermint 2100 8722 250 5500 272 4250 75 0.16
Zilliqa 279 9955 58 1100 22 458900 9276 0.24

Table 2: Adoption score calculation video tutorials, 1 point for tutorials for building a complete
application and 1 point in case the platform provides inter-
Platform Industrial Google Adoption active tools to learn to program it.
projects Scholar score Let’s now analyze the platforms’ documentation one by one:
papers Ethereum presents official documentation that defines the
Ethereum 18 84100 1 fundamental concepts, different tutorials are proposed, videos
Hyperledger 13 21400 0.49 are available, tutorials for developing complete projects are
Fabric detailed and some tools are proposed to learn in an interac-
Hyperledger 0 2810 0.02 tive way such as Tenderly, ChainIDE, and Replit. As a result,
Sawtooth it has a score of 5.
Hyperledger 0 973 0.01 Hyperledger Fabric, Hyperledger Sawtooth, R3 Corda, Quo-
Burrow rum, Stellar, MultiChain, and Tendermint documentations
Hyperledger 1 2490 0.04 do not include videos or interactive tools. They have a score
Indy of 3.
Hyperledger 0 959 0.01 Hyperledger Burrow and Indy have a score of 1 because
Iroha they present very rapid documentation. Hyperledger Iroha,
R3 Corda 3 5420 0.12 however, is given a score of 4 since all the criteria are met
Quorum 1 6150 0.06 except the last one: no interactive tools are proposed. Finally,
Stellar 0 7280 0.04 Zilliqa has a score of 2 since only basic tutorials are provided.
MultiChain 0 4450 0.03 • Security: In terms of security, we take into consideration
Tendermint 0 3000 0.02 multiple criteria. Similar to the way we calculated the score
Zilliqa 0 544 0 for documentation, the security score is calculated by giving
1 point to each aspect.
1 point for access management: private blockchains, which
do not allow access to non-authenticated peers are given 1
of the documentation, the security level, the maintenance, and the point, whereas public blockchains that allow any entity to ac-
proposed roadmap. cess the blockchain such as Ethereum are not given this point.
• Documentation: The documentation score relies on the amount 1 other point is given if access rights are defined in order
and quality of the available documentation. We were inspired to propose new transactions or validate them, 1 additional
by the score proposed in [14] and we calculate it in such a point is assigned if it handles smart contracts which allows
way that one point is given for each of the following aspects: to impose regulations on transactions, and 2 last points if
1 point if the platform published simple official documen- the platform offers the possibility to create private chan-
tation, 1 point if a written tutorial is presented, 1 point for nels between the nodes so that some transactions are only
ICCCP’23, December 2023, Lingshui, Hainan, China Riahi et al.

Table 3: Overall information about the platforms

Platform Application Initial Block- Consensus Smart Through- Popula- Adoption Maturity
areas GitHub chain Protocol Contract De- put (TPS) rity score score score
com- Type velopment
mit Language
Ethereum B2C: FinTech, 2013 Public PoS Solidity 30 0.9 1 0.68
NFTs, Games
Hyperledger GovTec, Health- 2016 Private Solo, Kafka, Go, Java, 20000 0.48 0.49 0.88
Fabric care, B2B Sup- RAFT,PBFT Javascript,
ply Chains Typescript
Hyperledger Healthcare, En- 2016 Private PoET, Raft, Python, Go, 1000 0.25 0.02 0.56
Sawtooth terprise Appli- PBFT Javascript,
cations, Supply Rust, Java,
Chain C++, Swift
Hyperledger B2B projects 2015 Private Byzantine Java, 200 0.19 0.01 0.48
Burrow fault tolerant JavaScript
Tendermint and Go
Hyperledger B2B projects 2016 Private RBFT Java, 200 0.31 0.04 0.4
Indy JavaScript
and Go
Hyperledger B2B projects 2019 Private BFT Java, 20000 0.17 0.01 0.72
Iroha JavaScript
and Go
R3 Corda B2B:FinTech, 2015 Private Notary Pool, Kotlin, Java 100 0.27 0.12 0.6
Supply Chain, Raft
Healthcare
Quorum FinTech 2013 Private PoA,RAFT, Is- Solidity 150 0.28 0.06 0.64
tanbul BFT
Stellar Assets ex- 2014 Public FBA(Federated Java,JavaScript, 200 0.22 0.04 0.56
change Byzantine Go, Python...
Agreement)
MultiChain assets, pharma- 2018 Private PoW JavaScript 1000 0.06 0.03 0.64
ceutical indus-
try
Tendermint FinTech, Car- 2014 Private BFT Go, Pharo, 5000 0.16 0.02 0.6
bon Trading C++, Java,
Ruby, Python
Zilliqa FinTech, dApps 2017 Public BFT Scilla 3000 0.24 0 0,6

accessible to defined nodes since this is a very interesting 2000, 4 for TPSs between 2000 and 10000 and 5 for TPSs over
feature. 10000.
The security aspects of the platforms were given in section 4. The approximate TPS of each platform is given in figure 3
• Maintenace: In order to assign a score related to the main- which summarizes the overall important information on
tenance level, we checked the last commit on the GitHub each platform.
reposit for each platform. Note that they were visited at the • Roadmap: Concerning roadmaps, some platforms have a
beginning of September 2023. A score of 5 is given to plat- very detailed roadmap, others only have an overview and
forms updated during that month (from 1st to 8th September some do not have a roadmap at all. Score 5 is given to de-
2023), 4 to platforms updated from June to August 2023, 3 tailed roadmaps, 4 to roadmaps with fewer details, 3 to vague
to platforms updated from January to June 2023, and finally roadmaps, 2 to outdated roadmaps, and 1 if the platform has
a score of 1 for platforms that were not updated for 1 year no published roadmap.
(until August 2022). Ethereum, on its official website [3], proposes a detailed
• Scalability: A platform’s scalability can be measured by the roadmap including the coming changes, the reasons for these
number of transactions it can handle per second (TPS). We changes, and the delays. This is why we gave it a score of 5.
give a score of 1 to platforms with a TPS less than 100, 2 Hyperledger Fabric slightly presents its short-term roadmap
for TPSs between 100 and 200, 3 for TPSs between 200 and with defined objectives: it has a score of 4. Also, Hyperledger
A Comparative Study of Blockchain Development Platforms ICCCP’23, December 2023, Lingshui, Hainan, China

Indy has an archived roadmap [5] which has not been up- [7] 2023. Multichain Roadmap. Retrieved September 2023 from https://docs.
dated since October 2018. It has a score of 2. Since Corda multichain.org/getting-started/road-map
[8] 2023. Top 9 blockchain platforms to consider in 2024. Retrieved Septem-
has not updated its roadmap since 2021, a score of 2 is attrib- ber 2023 from https://www.techtarget.com/searchcio/feature/Top-9-blockchain-
uted. Quorum’s roadmap can be presented on an on-demand platforms-to-consider
[9] Andy Amoordon and Henrique Rocha. 2019. Presenting tendermint: Idiosyn-
webinar [6] registered in 2021, it has a score of 2. Stellar pro- crasies, weaknesses, and good practices. In 2019 IEEE International Workshop on
vides a recent roadmap but where the objectives are not very Blockchain Oriented Software Engineering (IWBOSE). IEEE, 44–49.
clear, it has a score of 4. Furthermore, Multichain rapidly [10] Manas Pratim Bhattacharya, Pavol Zavarsky, and Sergey Butakov. 2020. En-
hancing the security and privacy of self-sovereign identities on hyperledger
presents its roadmap [7] and has a score of 4. Tendermint indy blockchain. In 2020 International Symposium on Networks, Computers and
roadmap is outdated and has not been updated since 2021 Communications (ISNCC). IEEE, 1–7.
[1], it has a score of 2. Zilliqa’s roadmap dates from April [11] Stefano Bistarelli, Rocco De Nicola, Letterio Galletta, Cosimo Laneve, Ivan Mer-
canti, and Adele Veschetti. 2021. Stochastic modeling and analysis of the bitcoin
2022, we give it a score of 3 since we cannot consider it as protocol in the presence of block communication delays. Concurrency and Com-
outdated. Finally, we gave a score of 1 to the platforms that putation: Practice and Experience (2021), e6749.
[12] Lin Chen, Lei Xu, Nolan Shah, Zhimin Gao, Yang Lu, and Weidong Shi. 2017.
did not publish a roadmap, namely Hyperledger Sawtooth, On security analysis of proof-of-elapsed-time (poet). In Stabilization, Safety, and
Hyperledger Burrow, and Hyperledger Iroha. Security of Distributed Systems: 19th International Symposium, SSS 2017, Boston,
MA, USA, November 5–8, 2017, Proceedings 19. Springer, 282–297.
The overall information regarding each platform is summarized [13] Victor Clincy and Hossain Shahriar. 2019. Blockchain development platform com-
in table 3. It presents, for each platform, its important application parison. In 2019 IEEE 43rd annual computer software and applications conference
(COMPSAC), Vol. 1. IEEE, 922–923.
domains, the year of the initial Github commit, the privacy charac- [14] Collin Connors and Dilip Sarkar. 2022. Comparative Study of Blockchain Devel-
teristic (either it is public or private), the used consensus protocol, opment Platforms: Features and Applications. arXiv preprint arXiv:2210.01913
(2022).
the supported programming languages for developing smart con- [15] Yousef Hashem, Elmedin Zildzic, and Andrei Gurtov. 2021. Secure drone identi-
tracts, the throughput (presented as the number of transactions per fication with hyperledger Iroha. In Proceedings of the 11th ACM Symposium on
second), and the three previously calculated scores for popularity, Design and Analysis of Intelligent Vehicular Networks and Applications. 11–18.
[16] Shereen Ismail, Hassan Reza, Hossein Kashani Zadeh, and Fartash Vasefi. 2023.
adoption, and maturity. A Blockchain-based IoT Security Solution Using Multichain. In 2023 IEEE 13th
After all, Ethereum remains the most popular and used platform Annual Computing and Communication Workshop and Conference (CCWC). IEEE,
overall, followed by Hyperledger, both having an interesting active 1105–1111.
[17] Alexander Ivanov, Yevhenii Babichenko, Hlib Kanunnikov, Paul Karpus, Leonid
community. Ethereum is mostly dedicated to B2C (Business-to- Foiu-Khatskevych, Roman Kravchenko, Kyrylo Gorokhovskyi, and Ievhen
consumer) solutions concerning payments and assets management. Nevmerzhitskyi. 2018. Technical comparison aspects of leading blockchain-based
platforms on key characteristics. (2018).
Hyperledger would rather be preferred for B2B projects when en- [18] Dodo Khan, Mehak Maqbool Memon, Manzoor Ahmed Hashmani, Filmann T
terprises are collaborating on a common project such as Supply Simpao, Anthony C Sales, and Neil Q Santillan. 2023. A CRITICAL REVIEW ON
Chains. It is ultimately up to the application developer to choose the BLOCKCHAIN FRAMEWORKS FOR DAPP. International Journal of Technology
Management and Information System 5, 1 (2023), 1–10.
most appropriate platform depending on the project requirements. [19] K Anitha Kumari, S Sangeetha, V Rajeevan, M Deva Dharshini, and T Haritha.
2022. Trade Management System Using R3 Corda Blockchain. In International
Conference on Intelligent Systems Design and Applications. Springer, 257–275.
6 CONCLUSION [20] Nitin Mittal, Srishty Pal, Anjali Joshi, Ashish Sharma, Sandeep Tayal, and Yogesh
In this paper, we presented a comprehensive comparative analysis Sharma. 2021. Comparative analysis of various platforms of blockchain. Smart
and Sustainable Intelligent Systems (2021), 323–340.
of the 12 most used blockchain development platforms. We pre- [21] Nishara Nizamuddin, Khaled Salah, M Ajmal Azad, Junaid Arshad, and MH
sented each one’s characteristics and proposed scores calculation Rehman. 2019. Decentralized document version control using ethereum
blockchain and IPFS. Computers & Electrical Engineering 76 (2019), 183–197.
for assessing their popularity, adoption, and maturity. Even though [22] Sandeep Kumar Panda and Suresh Chandra Satapathy. 2021. An investigation
Ethereum and Hyperledger Fabric have the highest scores overall into smart contract deployment on Ethereum platform using Web3. js and solidity
based on our study, a deeper analysis related to the application using blockchain. In Data Engineering and Intelligent Computing: Proceedings of
ICICC 2020. Springer, 549–561.
needs is required. [23] Julien Polge, Jérémy Robert, and Yves Le Traon. 2021. Permissioned blockchain
As a future work, it would be interesting to implement a project frameworks in the industry: A comparison. Ict Express 7, 2 (2021), 229–233.
using each of the 12 platforms to be able to analyze the difficulties [24] Swathi Punathumkandi and Venkatesan Meenakshi. 2020. A deep dive into
Hyperledger. Blockchain and Machine Learning for e-Healthcare Systems (2020),
encountered in the development process as well as the facilities 85.
provided and the technical characteristics which would be useful [25] Kenza Riahi, Mohamed-el-Amine Brahmia, Abdelhafid Abouaissa, and Lhassane
Idoumghar. 2022. APBFT: An Adaptive PBFT Consensus for Private Blockchains.
for future developers. In GLOBECOM 2022 - 2022 IEEE Global Communications Conference. 1788–1793.
https://doi.org/10.1109/GLOBECOM48099.2022.10001568
[26] Chinmay Saraf and Siddharth Sabadra. 2018. Blockchain platforms: A com-
REFERENCES pendium. In 2018 IEEE International Conference on Innovative Research and Devel-
[1] 2021. Tendermint Roadmap. Retrieved September 2023 from https://github.com/ opment (ICIRD). IEEE, 1–6.
bdengine/bde-tendermint/blob/master/docs/roadmap/roadmap.md [27] Fabian Schär. 2021. Decentralized finance: On blockchain-and smart contract-
[2] 2023. The CMMI Ecosystem. Retrieved September 2023 from https://cmmiinstitute. based financial markets. FRB of St. Louis Review (2021).
com/ [28] A Secure. 2018. The zilliqa project: A secure, scalable blockchain platform. (2018).
[3] 2023. Ethereum Roadmap. Retrieved September 2023 from https://ethereum.org/
en/roadmap
[4] 2023. Forbes Blockchain 50 2022. Retrieved September 2023 from
https://www.forbes.com/sites/michaeldelcastillo/2022/02/08/forbes-
blockchain-50-2022/?sh=6ebe51d831c6
[5] 2023. Hyperledger Indy Roadmap. Retrieved September 2023 from https://wiki-
archive.hyperledger.org/projects/indy/roadmap
[6] 2023. Introducing ConsenSys Quorum. Retrieved September 2023 from https:
//insights.consensys.io/introducing-consensys-quorum

View publication stats

You might also like