Journal of Computer Science

Original Research Paper

Application of Graph Theory Concepts in Computer

Networks and its Suitability for the Resource Provisioning
Issues in Cloud Computing-A Review
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy

Department of Computer Science and Engineering, Kalasalingam University, Krishnankoil, Tamilnadu, India

Article history Abstract: Cloud computing, a kind of web service provisioning model,
Received: 23-08-2017 provides immense benefits over traditional IT service environments with
Revised: 30-11-2017 the help of virtualization technology. As cloud computing is not a fully
Accepted: 18-01-2018 matured paradigm, it poses many open issues to be addressed. The key
Corresponding Author:
research problem in cloud computing is efficient resource provisioning
Kanniga Devi Rangaswamy which is due to its complex and distributed architecture. Graph-based
Department of Computer representations of complex networks provide simpler views and graph
Science and Engineering, theoretical techniques provide simpler solutions for lot of issues inherent in
Kalasalingam University, networks. Hence, this paper begins with exploration of graph theory
Krishnankoil, Tamilnadu, India applications in computer networks with specific focus on graph theory
Email: [email protected] applications in cloud computing. This work pays attention to basic resource
provisioning problems arise in cloud computing environments and presents
some conceptual graph theoretical suggestions to address these issues.

Keywords: Computer Network, Cloud Computing, Graph Theory,

Resource Provisioning

Introduction computing. First, it starts with graph theory applications

in various areas in computer networks and then explores
The next step in the evolution of distributed its suitability to address the resource provisioning issues
computing is cloud computing. It inherits existing of cloud. Graph theory is part of discrete mathematics
distributed computing models such as grid computing,
and useful structure to model relationship between
utility computing and adds additional flavour namely
objects. Graph theory mainly finds its applications in
virtualization. Large-scale processing and storage of data
network modeling, biology, electrical network,
are very much simplified with the advent of cost-
effective cloud computing solutions. The Cloud Data computational algorithms and scheduling. Graph
Center (CDC) (Wu and Buyya, 2015) is very complex theoretical techniques are highly used by computer
with resources distributed globally leading to several science applications especially in modeling and routing
issues. The authors of (Zhang et al., 2010; Smith et al., in networks. Representing a problem as a graph can
2009; Marston et al., 2011; Rimal et al., 2009; Shahzad, provide a different point of view and makes a problem
2014) addressed several key and fundamental cloud much simpler. It provides tools for solving the problem
computing issues like resource provisioning, security, and set of techniques for analysing graphs.
privacy, energy and interoperability, but this list is not This work has two main parts. The first part gives an
limited. From the perspective of cloud service provider overview of graph theory applications in computer
and cloud service consumer, these issues provide networks. As cloud data center has a set of
different pictures. While cloud computing provides interconnected systems, graph theoretical solutions on
opportunities to migrate the IT business services online, computer networks can well be applied on cloud with
these key issues need to be resolved before it is accepted suitable modifications to address its issues. The second
as a successful business model. part gives an overview of graph theory applications in
This article identifies the opportunities of graph theory cloud. The main idea behind this work is to find out the
(Bondy and Murty, 1976) based solutions for the scope for applicability of graph theory to address
resource provisioning issues inherent in cloud resource provisioning issues in cloud.

© 2018 Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy. This open access article is distributed under a Creative
Commons Attribution (CC-BY) 3.0 license.
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy / Journal of Computer Science 2018, 14 (2): 163.172
DOI: 10.3844/jcssp.2018.163.172

Key Research Issues in Cloud Computing Server consolidation are the major threatening factor for
fault tolerance as they incur service downtime.
Resource Provisioning
Interoperability
The authors of (Meng et al., 2010; Buyya et al.,
2011; Vecchiola et al., 2012; Javadi et al., 2012; The authors of (Govindrajan and Lakshmanan, 2010;
Warneke and Kao, 2011) proposed that unlike traditional Loutas et al., 2011) stated that interoperability of
resource provisioning where resources are provisioned as heterogeneous cloud platforms are difficult because they
it is, cloud demands efficient resource provisioning use distinct hypervisor and VM technologies. The
algorithms to provision virtualized resources to meet SLA platforms also use various security standards and
requirements. Virtualized data centers are envisioned to management interfaces. Multiple vendors with
provide better management flexibility, lower cost, different product standards poses challenges for
scalability, better resources utilization and energy interoperability. Cloud adoption will be stopped if
efficiency, but virtualization is not an easy task to do. there is not a good way of integrating data and
applications across clouds; hence a unified cloud
Security interface and open standards need to be developed.
The authors of (Denz and Taylor, 2013; Ward and Energy
Barker, 2014; Mahajan et al., 2011; Subashini and
Kavitha, 2011; Zissis and Lekkas, 2012) discussed that The authors of (Beloglazov et al., 2011; Guo and
as users and corporate information reside on third-party Fang, 2013; Dayarathna et al., 2016; Beloglazov and
systems, no one can guarantee how secure the data are. It Buyya, 2010; Lee and Zomaya, 2010) stated that
is prone to leakage of information and attack. Security is increasing demand for computational power, leads to
a primary issue which should be handled by all the cloud setting up large-scale data centers. On the other side, the
service providers to retain their business in the market. power consumption of these large-scale data centers is
They should take steps to protect data and its privacy. enormous. Hence, design of energy efficient hardware
Five most representative security and privacy attributes and intelligent resource management techniques is
required. Due to enormous power consumption, carbon
are confidentiality, integrity, availability, accountability
dioxide (CO2) emission is also more contributing to the
and privacy-preservability.
greenhouse effect. Hence number of practices need to be
Cost applied to achieve energy efficiency, such as
improvement of applications’ algorithms, energy
The authors of (Sahni and Vidyarthi, 2015; Tsai and efficient hardware and energy-efficient resource
Liao, 2015; Waibel et al., 2016; Hadji et al., 2016) management strategies on a virtualized data center.
proposed that the public cloud offers pay per use, which Out of all these above mentioned key issues, this
can provide low-cost options for short-term projects. work pays much attention to resource provisioning issue
Still, for long-term use, enterprise IT organizations may and application of graph theory on it.
be better off making a capital investment to purchase
additional hardware and software. Enterprises need to Resource Provisioning in Cloud
conduct a break-even analysis to determine whether a In cloud computing, resource provisioning (Meng et al.,
public or private cloud would be more cost-effective for 2010; Buyya et al., 2011; Vecchiola et al., 2012;
them. From the perspective of provider, they are Javadi et al., 2012; Warneke and Kao, 2011) is the
interested in customer satisfaction and generating allocation of a cloud data center resources to a user.
revenue out of their services. From the perspective of When cloud data center accepts requests for hardware or
consumer, they are interested in cost-effective solutions. software resources, it must create and provision them as
To balance these two points, cost-effective cloud virtualized resources. It means the monitoring, dynamic
solutions need to be developed. selection/scheduling, deployment/placement and
Reliability management/load balancing of software and hardware
resources for ensuring Service Level Agreement (SLA).
The authors of (Vishwanath and Nagappan, 2010; The SLA is an agreement between the cloud service
Poola et al., 2014; Di and Wang, 2013; Zheng et al., provider and the user on guaranteeing Quality of Service
2012) stated that due to high system complexity and (QoS). The provisioning can be done in several different
distributed structure, even carefully engineered data ways. In particular, this work addresses the following
centers are subject to a large number of failures. Fault aspects of resource provisioning from CDC:
tolerant systems should be built to address reliability
concerns. Because of the abstraction nature of cloud • Efficient monitoring for provisioning CDC resources
environment, there arise a need to develop new or extend • Optimal VM placement and migration in CDC for
traditional fault tolerant approaches. VM migration and energy-efficient resource provisioning

164
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy / Journal of Computer Science 2018, 14 (2): 163.172
DOI: 10.3844/jcssp.2018.163.172

• Proper locating of CDCs and allocation of CDCs to perform load balancing. But there is a trade-off always
the source of requests exits between choosing the closest cloud data center and
• Clustering distributed CDCs for faster server balancing the load of cloud data center. Sometimes a
provisioning cloud data center closer to the user location may be in
• Uniform assignment of clients to CDC servers overloaded condition, during this case, the requests will
• Traffic-aware VM migration to load balance cloud be routed to a distant cloud data center which is capable
servers of handling the requests. Hence, there arise a need to
deal with this trade-off which may consider both
Efficient Monitoring for Provisioning CDC Resources proximity and load at the same time.

Task scheduling, load balancing are complicated in Traffic-Aware VM Migration to Load Balance
cloud computing environment due to its abstract Cloud Servers
heterogeneous architecture, dynamic behaviour and Upon receiving the load information, cloud broker
resource heterogeneity. Monitoring of resources is required must invoke load balancing procedure to distribute the
before performing scheduling and load balancing. load uniform across the hosts in the CDC. It can be done
Optimal VM Placement and Migration in CDC for by migrating some of the VMs from overloaded hosts to
Energy-Efficient Resource Provisioning underloaded hosts considering only server-side
constraints. Network-side constraints also need to be
Keeping lot of PMs and VMs running in the considered to enhance the performance of CDCs.
datacenter consumes more energy, leading to higher
operating costs. Hence identifying physical machines Graph Theory Applications in Computer
with least load and migrating its load to some other
physical machines and then shutting them down saves
Networks
energy. Conservation of energy may be better achieved This section summary of some of the works, which
through optimal placement of VMs on the PMs and applied graph theory in various types of computer network.
performing VM migrations, so that energy consumption Table 1 lists some of the possible graph theory
may be maintained at desirable level. applications in various types of network. Since cloud is a
kind of network, these traditional graph theoretical
Proper Locating of CDCs and Allocation of CDCs
techniques can be analyzed for their suitability to address
to the Source of Requests resource provisioning issues in cloud and also presents
The requests for the CDC services can come from some conceptual suggestions for it.
different parts of the world. The term source of
requests/clients denote the users who make requests to Graph Theory Applications in Cloud
various cloud data center services. The distance between Computing
the cloud data center and the source of requests is a
major factor influencing the quality of service in terms of This section provides summary of works, which
response time and latency. Cloud data center allocation applied graph theory in cloud.
is one of the major issues in cloud computing. An The author of (Çatalyürek et al., 2011) proposed a
efficient allocation of cloud data center to the source of heuristic based on hypergraph and its partitioning for
requests may improve the quality of services. optimizing scientific workflow execution. The author of
(Dai et al., 2009) modeled the reliability of the cloud
Clustering Distributed CDCs for Faster Server services. This paper models network failure, overflow
Provisioning failure, timeout failure, resource missing failure,
Normally cloud data centers are distributed across the hardware failure, software failure and database failure
failures using Markov models, Queuing Theory and
world to increase the availability of services by remote
Graph Theory. The author of (Verbelen et al., 2013)
mirroring, replication which are the kind of redundancy
investigated how to optimally deploy software
mechanisms. It is distributed mainly for disaster
applications on the offered infrastructure in the cloud, by
recovery. Clustering region-wise deployed cloud data
minimizing the network usage. Especially in the context
centers will provide rapid responses. of mobile computing. They designed and evaluated
Uniform assignment of clients to CDC servers graph partitioning algorithms that allocate software
components to machines in the cloud while minimizing
In a distributed cloud data center environment, load the required bandwidth. The author of (Li et al., 2012)
balancing techniques direct the requests to the closest proposed CAM, a cloud platform that provides an
source or to the source with the most able capacity to innovative resource scheduler particularly designed for
serve the request. Variety of algorithms are used to hosting MapReduce applications in the cloud.

165
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy / Journal of Computer Science 2018, 14 (2): 163.172
DOI: 10.3844/jcssp.2018.163.172

Table 1: Graph theory in computer networks

Author Graph theory technique Issue addressed Type of network
Wang et al. (2006; Graph coloring Interference reduction, Interference Aware Wireless networks-Mobile
Tamura and Nakano, 2011; TDMA Link scheduling, Job Scheduling/ Adhoc Networks and Sensor
Tamura et al., 2008) Assignment problems, Resource allocation networks
Alzoubi et al. (2002; Dominating set and Routing, fault tolerance, energy- efficiency,
Scheideler et al., 2008; Connected dominating set delivery delay reduction, connectedness,
Wu and Li, 1999; Erciyes et al., Virtual backbone construction for efficient
2007; Liu et al., 2007; routing, overlay network construction,
Dai and Wu 2004; Search space reduction, Clustering nodes
Muhammad 2007)
Kawahigashi et al. (2005) Random graphs Connectivity, scalability, Routing,
Congestion handling, Modeling the network
Meghanathan (2012) Shortest path algorithms Route computation for communication
Wang and Li (2006) Spanner Minimum power assignment
Dong et al. (2010) Topological graph Network Coverage problem
Newsome and Song (2003) Graph embedding Routing
Prasanna et al. (2014) Graph labeling Fast communication via radio labeling
Bose et al. (2001) Gabriel graph, unit graph Routing with guaranteed delivery
Xie et al. (2010) DAG Workflow modeling, Loop-free routing
Xie et al. (2014) DAG Task Scheudling Network embedded systems
Nakayama and Koide (2013) Network flow Flow optimization Mesh network
Fatmi and Pan (2014) Tree Data center/ network modeling Datacenter
Bunke et al. (2006) Graph matching Abnormal event detection in network General network
by graph comparison
Jana and Naik (2009) Spanning and minimum Loop-free connectedness, clustering
spanning tree algorithms
Wang et al. (2012) Graph traversal Searching an object General network query
optimization
Maldeniya et al. (2013; Graph partitioning Clustering, faster communication Real road networks,
Tang et al., 2008) World Wide Web
Dai et al. (2007; 2009) Virtualized graph model Grid service reliability evaluation Grid
Venugopal and Buyya (2006) Set covering problem Mapping applications with data sets
Xiao and Parhami (2003) Cayley digraphs Design of scalable Interconnection networks
Bentley (1979) KD-Tree Multi-dimensional search Database
Potluri and Singh (2012) Capacitated dominating set Minimum capacitated dominating Wireless networks
set construction
Bar-Ilan et al. (1992) Facility location problem Locating centers General facility

Table 2: Graph theory in cloud

Author Graph theory technique Issue addressed Limitation
Çatalyürek et al. Hypergraph Modeling cloud network Scheduling is done in two-phases which increases makespan. If
(2011) done in single phase, it may minimize the makespan of the
workflow. This work didn't consider dynamic workflows whose
execution pattern changes over time as well.
Dai et al. (2009) Spanning tree Cloud service reliability Cloud service reliability model and evaluation algorithm have
not been validated by simulation and real-life data. They have
proposed only theoretical model.
Verbelen et al. Graph partitioning Software deployment in This work considered only minimizing the bandwidth between
(2013) mobile cloud software components. It didn't minimize the execution time
of tasks. This work provide scope for integrating their algorithms
to address energy consumption objective.
Li et al. (2012) Flow network algorithms Optimizing MapReduce Validation of min-cost flow model for data and VM placement
-Ford-Fulkerson performance in cloud is to be done in larger real VM clusters.
Binz et al. (2012) Topological graph theory Topology management Enterprise Topology Graph (ETG) is to be built as reusable
in cloud building blocks to address specific problem domains with
broad set of basic operations.
Bansal et al. (2011) Graph mapping Congestion minimization Only mapping of two classes of workloads, namely depth-d
in cloud network trees and complete graph are considered.
Chan et al. (2009) Predicate-based graph Modeling and Testing This model supports only stateless atomic operations expressed
cloud in context free grammar. Some other attributes such as scalability,
exception handling, dynamic binding, service interactions can also
be considered to make the model valuable.
Peng et al. (2015) Graph cut theory Virtual Machine (VM) Time taken for VM clustering is a valuable parameter, which
clustering in cloud is not mentioned. Not mentioned the applicability of
Zegzhda and Graph model Security model for VM hypervisor security model on the specific type of hypervisor
Nikolsky (2014) hypervisor in cloud architectures, namely hosted or bare-metal.

166
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy / Journal of Computer Science 2018, 14 (2): 163.172
DOI: 10.3844/jcssp.2018.163.172

CAM uses a flow-network-based algorithm that is able which increases network/message overhead. To
to optimize MapReduce performance under the resolve this, this work suggests to represent cloud as a
specified constraints. The author of (Binz et al., 2012) graph and construct a minimum dominating set
proposed Enterprise Topology Graphs (ETG) as formal structure, which can be used for monitoring resource
model to describe an enterprise topology. The author of status. The minimum dominating set structure can
(Bansal et al., 2011) represented the physical network of minimize the number of message updates made to the
the cloud as a graph considering the objective of monitor and minimize the update time compared to
minimizing the congestion. The author of (Chan et al., traditional monitoring methods.
2009) formulated a computing cloud as a kind of graph,
a computing resource such as services or intellectual How to Perform Optimal VM Placement and
property access rights as an attribute of a graph node and Migration in CDC for Energy-Efficient Resource
the use of the resource as a predicate on an edge of the Provisioning?
graph. It also proposes to model cloud computation as a As VM placement has direct impact on the
set of paths in a subgraph of the cloud such that every
performance of the CDC, this work suggests to optimize
edge contains a predicate that is evaluated to be true.
VM placement and VM migration for energy-efficient
Finally, it presents a set of algorithms to compose cloud
resource provisioning. For optimal VM placement, it
computations and model-based testing criteria to test
cloud applications. The author of (Peng et al., 2015) suggests to construct KD-tree structure of cloud servers
constructed a method for VM cluster by energy and VMs will be placed on best-fit cloud servers quickly.
minimization based on graph theory. Then, they changed This KD-tree is a useful structure when search involves a
deployment of VM cluster into maximum flow minimum multidimensional search key and it is known as
cut problem and finally cut formed for VM cluster. The associative search. KD-tree is defined as a tree for storing
author of (Zegzhda and Nikolsky, 2014) described a the values or objects in multidimensional space. They
formal security model for virtual machine hypervisors in scale well in high dimensions. All the resource dimensions
cloud systems based on the graph theory. namely CPU, RAM, storage and bandwidth of cloud
From Table 2, it is observed that only a very minimal servers and VM can well be represented using KD-tree
literature is available on graph theory applications in structure. Moreover, this structure helps in identifying
cloud and most of these works overlooked its application least loaded PMs easily so that VMs running on it can be
on resource provisioning. Thus, it opens lot of migrated to other cloud servers and powered off to
opportunities to apply graph theoretical techniques to conserve energy. The author of (Mandal and Khilar, 2013)
address resource provisioning issues in cloud. constructed one dimensional Binary Search Tree, which
The following section presents some conceptual considers only single resource dimension, CPU.
suggestions for applying graph theory for resource
provisioning issues in cloud computing. How to Ensure Proper Locating of CDCs and
Allocation of CDCs to the Source of Requests?
Some Conceptual Suggestions for the This suggests to model cloud data center network as a
Suitability of Graph Theory Techniques for graph and use facility location problem to locate CDCs
Addressing Resource Provisioning in Cloud in appropriate locations to serve the source of requests.
Limiting the distance between cloud data centers and the
This section considers some of the graph theoretical source of requests leads to faster service provisioning.
techniques from Table 1 and suggests the following The author of (Doyle et al., 2013) has proposed the
aspects of resource provisioning in cloud and the source of requests assignment to the closest cloud data
suitability of such graph theoretical models and center to reduce the carbon emission but they modeled
techniques to address its issues. cloud data center as a complete graph which is
How to Perform Efficient Monitoring for Provisioning unrealistic. They modelled both the networking and
computational components of the infrastructure as a
CDC Resources?
graph and proposed a system which utilizes Voronoi
Monitoring the complex, distributed cloud partitions to determine how source requests to be routed
architecture for tracking resource status is a challenging to appropriate data center based on the relative priorities
task. Monitoring ensures the availability of resources of the cloud operator for latency purposes. The author of
before performing successful provisioning. Generally, a (Bar-Ilan et al., 1992) provided solutions for facility
central monitor keeps track of the resources in cloud location problems. They have considered distributing the
(Galstad, 2016). clients to centers as balanced as possible, but they have
The monitor has to query all the resources in the overlooked the distance between clients and centers,
network periodically to get their availability status, which is also essential for faster service provisioning.

167
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy / Journal of Computer Science 2018, 14 (2): 163.172
DOI: 10.3844/jcssp.2018.163.172

How to Cluster Distributed CDCs for Faster Server path. In cloud, some k number of edge-disjoint paths can
Provisioning? be computed between overloaded and underloaded
servers and can migrate each VM from overloaded
This suggests to model the cloud environment as a server to underloaded server on one of these k paths to
graph and cluster distributed CDCs in order to enable perform fast and traffic-aware VM migrations.
faster service provisioning to the clients. Many works
have considered graph-based K-Means (Galluccio et al.,
2012) and K-Spanning tree (Zhou et al., 2011) clustering
Conclusion and Future Work
methods to cluster a network. In all these approaches the In this study, an analysis is done on the application of
number of clusters K to be created should be known in graph theory concepts in computer networks and its
advance, but determining K is impractical in case of suitability to address resource provisioning issues in
distributed CDCs. Hence, this work suggests to use cloud. In future, we would like to analyze its scope in
dominating set based clustering, where each dominating other research areas of cloud computing.
node acts as a cluster head and neighbors are connected
to these cluster heads which makes service provisioning Acknowledgment
fast. The number of clusters (K) to be created is
determined based on size of the graph. So, it leads to a The authors would like to thank the support rendered by
non-constrained clustering algorithm. Kalasalingam University to carry out this review work.

How to Ensure Uniform Assignment of Clients to Author’s Contributions

CDC Servers?
Kanniga Devi Rangaswamy: She devised the main
This work suggests to use capacitated dominating set conceptual ideas and developed the review framework.
concept (Potluri and Singh, 2012) to assign clients to She contributed to part of the literature collection and
CDC servers. The capacitated dominating set is the drafted the manuscript in consultation with
problem of finding a dominating set with the additional Murugaboopathi Gurusamy.
constraint that the nodes dominated do not exceed the Murugaboopathi Gurusamy: He supervised this
capacity of the dominating node. The capacity can be review work. He contributed to part of the literature
uniform across all nodes or variable. Hence, a uniform collection and to the formalism of review work. He
capacity can be assigned to a homogeneous CDC provided critical feedback and helped shape the research
environment and a variable capacity can be assigned to a and manuscript.
heterogeneous CDC environment. For client assignment
to CDCs, exiting methods have used Round Robin Ethics
(Radojević and Žagar, 2011) and its variant approaches,
which may not guarantee uniformity all the times. This research manuscript is original and the
manuscript is not published elsewhere. The corresponding
How to Efficiently Perform Traffic-Aware VM author ensure that all authors have read and approved the
Migration to Load Balance Cloud Servers? manuscript and no ethical issues involved.
Network flow problem (Nakayama and Koide, 2013)
in the form of Maximum flow and Minimum cut References
problem finds its application in various fields ranging Alzoubi, K.M., P.J. Wan and O. Frieder, 2002. Message-
from checking network connectivity, network reliability, optimal connected dominating sets in mobile ad hoc
namely a few. networks. Proceedings of the 3rd ACM International
This work suggests to use network flow problem by Symposium on Mobile Ad Hoc Networking and
converting cloud into a flow network and to perform Computing, Jun. 09-11, ACM, Lausanne,
traffic-aware live VM migration in order to load balance Switzerland, pp: 157-164.
the cloud servers. Most of the existing benchmarked VM DOI: 10.1145/513800.513820
migration approaches (Khanna et al., 2006; Wood et al., Bansal, N., K.W. Lee, V. Nagarajan and M. Zafer, 2011.
2007) consider only host-side resource constraints with Minimum congestion mapping in a cloud.
little consideration on network-side constraints. A good Proceedings of the 30th Annual ACM SIGACT-
VM migration algorithm which considers both server- SIGOPS Symposium on Principles of Distributed
side and network-side constraints can greatly improve Computing, Jun. 06-08, ACM, San Jose, California,
their performance. This can be done on both live USA, pp: 267-276. DOI: 10.1145/1993806.1993854
migration or cold migration scenarios. Bar-Ilan, J., G. Kortzars and D. Pelag, 1992. How to
In network flow problem, a set of path in a graph G is allocate network centers. J. Algorithms, 15: 385-415.
edge-disjoint if each edge in G appears in at most one DOI: 10.1006/jagm.1993.1047

168
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy / Journal of Computer Science 2018, 14 (2): 163.172
DOI: 10.3844/jcssp.2018.163.172

Beloglazov, A. and R. Buyya, 2010. Energy efficient Dai, F. and J. Wu, 2004. An extended localized algorithm
resource management in virtualized cloud data for connected dominating set formation in ad hoc
centers. Proceedings of the 10th IEEE/ACM wireless networks. IEEE Trans. Parallel Distributed
International Conference on Cluster, Cloud and Grid Syst., 15: 908-920. DOI: 10.1109/TPDS.2004.48
Computing, May 17-20, IEEE Xplore Press, Dai, Y.S., B. Yang, J. Dongarra and G. Zhang, 2009.
Melbourne, VIC, Australia, pp: 826-831. Cloud service reliability: Modeling and analysis.
DOI: 10.1109/CCGRID.2010.46 Innovative Computing Laboratory, Department of
Beloglazov, A., R. Buyya, Y.C. Lee and A. Zomaya, Electrical Engineering and Computer Science,
2011. A taxonomy and survey of energy-efficient University of Tennessee, Knoxville.
data centers and cloud computing systems. Adv. Dai, Y.S., Y. Pan and X. Zou, 2007. A hierarchical
Comput., 82: 47-111. modeling and analysis for grid service reliability.
DOI: 10.1016/B978-0-12-385512-1.00003-7 IEEE Trans. Comput., 56: 681-691.
Bentley, J.L., 1979. Multidimensional binary search DOI: 10.1109/TC.2007.1034
trees in database applications. IEEE Trans. Software Dayarathna, M., Y. Wen and R. Fan, 2016. Data center
Eng., 5: 333-340. DOI: 10.1109/TSE.1979.234200 energy consumption modeling: A survey. IEEE
Binz, T., C. Fehling, F. Leymann, A. Nowak and D. Commun. Surveys Tutorials, 18: 732-794.
Schumm, 2012. Formalizing the cloud through DOI: 10.1109/COMST.2015.2481183
enterprise topology graphs. Proceedings of the IEEE Denz, R. and S. Taylor, 2013. A survey on securing the
5th International Conference on Cloud Computing, virtual cloud. J. Cloud Comput.: Adv. Syst. Applic.,
Jun. 24-29, IEEE Xplore Press, Honolulu, HI, USA, 2: 17-17. DOI: 10.1186/2192-113X-2-17
pp: 742-749. DOI: 10.1109/CLOUD.2012.143 Di, S. and C.L. Wang, 2013. Error-tolerant resource
Bondy, J.A. and U.S.R. Murty, 1976. Graph Theory with allocation and payment minimization for cloud
Applications. 1st Edn., North Holland, New York, system. IEEE Trans. Parallel Distributed Syst., 24:
ISBN: 0444194517, pp: 264. 1097-1106. DOI: 10.1109/TPDS.2012.309
Bose, P., P. Morin, I. Stojmenović and J. Urrutia, 2001. Dong, D., Y. Liu, K. Liu and X. Liao, 2010. Distributed
Routing with guaranteed delivery in ad hoc wireless coverage in wireless ad hoc and sensor networks
networks. Wireless Netw., 7: 609-616. by topological graph approaches. Proceedings of
DOI: 10.1006/jagm.1993.1047 the IEEE 30th International Conference on
Bunke, H., P. Dickinson, A. Humm, C. Irniger and M. Distributed Computing Systems, Jun. 21-25, IEEE
Kraetzl, 2006. Computer network monitoring and Xplore Press, Genova, Italy, pp: 1417-1428.
abnormal event detection using graph matching and DOI: 10.1109/ICDCS.2010.9
multidimensional scaling. Proceedings of the 6th Doyle, J., R. Shorten and D. O’Mahonym, 2013. Stratus:
Industrial Conference on Data Mining Conference Load balancing the cloud for carbon emissions
on Advances in Data Mining: Applications in control. IEEE Trans. Cloud Comput., 1: 1-1.
Medicine, Web Mining, Marketing, Image and DOI: 10.1109/TCC.2013.4
Signal Mining, Jul. 14-15, Springer, Leipzig, Erciyes, K., O. Dagdeviren, D. Cokuslu and D. Ozsoyeller,
Germany, pp: 576-590. DOI: 10.1007/11790853_45 2007. Graph theoretic clustering algorithms in mobile
Buyya, R., S.K. Garg and R.N. Calheiros, 2011. SLA- ad hoc networks and wireless sensor networks survey.
oriented resource provisioning for cloud Applied Comput. Math., 6: 162-180.
computing: Challenges, architecture and solutions. Fatmi, O. and D. Pan, 2014. Distributed multipath routing
Proceedings of the International Conference on for data center networks based on stochastic traffic
Cloud and Service Computing, Dec. 12-14, IEEE modeling. Proceedings of the IEEE 11th International
Xplore Press, Hong Kong, China, pp: 1-10. Conference on Networking, Sensing and Control,
DOI: 10.1109/CSC.2011.6138522 Apr. 7-9, IEEE Xplore Press, Miami, FL, USA,
Çatalyürek, U.V., K. Kaya and B. Uçar, 2011. Integrated pp: 536-541. DOI: 10.1109/ICNSC.2014.6819683
data placement and task assignment for scientific Galluccio, L., O. Michel, P. Comon and A.O. Hero, 2012.
workflows in clouds. Proceedings of the 4th Graph based K-means clustering. Signal Process., 92:
1970-1984. DOI: 10.1016/j.sigpro.2011.12.009
International Workshop on Data-Intensive Distributed
Galstad, E., 2016. Nagios NRPE Documentation.
Computing, Jun. 08-08, ACM, San Jose, California,
Govindrajan, A. and Lakshmanan, 2010. Overview of
USA, pp: 45-54. DOI: 10.1145/1996014.1996022 Cloud Standards Cloud Computing. 1st Edn.,
Chan, W.K., L. Mei and Z. Zhang, 2009. Modeling and Springer London.
testing of cloud applications. Proceedings of the Guo, Y. and Y. Fang, 2013. Electricity cost saving
IEEE Asia-Pacific Services Computing Conference, strategy in data centers by using energy storage.
Dec. 7-11, IEEE Xplore Press, Singapore, pp: 111-118. IEEE Trans. Parallel Distributed Syst., 24: 1149-1160.
DOI: 10.1109/APSCC.2009.5394131 DOI: 10.1109/TPDS.2012.201

169
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy / Journal of Computer Science 2018, 14 (2): 163.172
DOI: 10.3844/jcssp.2018.163.172

Hadji, M., B. Aupeti and D. Zeghlache, 2016. Cost- Mahajan, P., S. Setty, S. Lee, A. Clement and L.
efficient algorithms for critical resource Alvisi et al., 2011. Depot: Cloud storage with
allocation in cloud federations. Proceedings of the minimal trust. ACM Trans. Comput. Syst.
5th IEEE International Conference on Cloud DOI: 10.1145/2063509.2063512
Networking, Oct. 3-5, IEEE Xplore Press, pp: 1-6. Maldeniya, S.L., A.S. Atukorale and W.W. Vithanage,
DOI: 10.1109/CloudNet.2016.11 2013. Network data classification using graph
Jana, P.K. and A. Naik, 2009. An efficient minimum partition. 19th Proceedings of the IEEE International
spanning tree based clustering algorithm. Conference on Networks, Dec. 11-13, IEEE Xplore
Proceedings of the International Conference on Press, Singapore, pp: 1-6.
Methods and Models in Computer Science, Dec. DOI: 10.1109/ICON.2013.6781952
14-15, IEEE Xplore Press, Delhi, India, pp: 1-5. Mandal, S.K. and P.M. Khilar, 2013. Efficient virtual
DOI: 10.1109/ICM2CS.2009.5397966 machine placement for on-demand access to
Javadi, B., J. Abawajy and R. Buyya, 2012. Failure- infrastructure resources in cloud computing. Int. J.
aware resource provisioning for hybrid Cloud Comput. Applic., 68: 6-11.
infrastructure. J. Parallel Distributed Comput., 72: Marston, S., Z. Li, S. Bandyopadhyay, J. Zhang and A.
1318-1331. DOI: 10.1016/j.jpdc.2012.06.012 Ghalsasi, 2011. Cloud computing-the business
Kawahigashi, H., Y. Terashima, N. Miyauchi and T. perspective. Decision Support Syst., 51: 176-189.
Nakakawaji, 2005. Modeling ad hoc sensor DOI: 10.1016/j.dss.2010.12.006
networks using random graph theory. Proceedings Meghanathan, N., 2012. Graph theory algorithms for
of the 2nd IEEE Consumer Communications and mobile ad hoc networks. Informatica, 36: 185-200.
Networking Conference, Jan. 6-6, IEEE Xplore DOI: 10.4018/978-1-4666-0080-5.ch004
Press, Las Vegas, NV, USA, pp: 104-109. Meng, X., C. Isci, J. Kephart, L. Zhang and E.
DOI: 10.1109/CCNC.2005.1405152 Bouillet et al., 2010. Efficient resource
Khanna, G., K. Beaty, G. Dhar and A. Kochu, 2006. provisioning in compute clouds via VM
Application performance management in virtualized multiplexing. Proceedings of the 7th International
Conference on Autonomic Computing, Jun. 07-11,
server environments. Proceedings of the 10th
ACM, Washington, DC, USA, pp: 11-20.
IEEE/IFIP Network Operations and Management
DOI: 10.1145/1809049.1809052
Symposium, Apr. 3-7, IEEE Xplore Press,
Muhammad, R.B., 2007. A distributed graph algorithm
Vancouver, BC, Canada, pp: 373-381. for geometric routing in ad hoc wireless networks. J.
DOI: 10.1109/NOMS.2006.1687567 Netw., 2: 50-57.
Lee, Y.C. and A.Y. Zomaya, 2010. Energy efficient Nakayama, K. and T. Koide, 2013. A decentralized
utilization of resources in cloud computing algorithm for network flow optimization in mesh
systems. J. Supercomput., 60: 268-280. networks. Proceedings of the IEEE Global
DOI: 10.1007/s11227-010-0421-3 Communications Conference, Dec. 9-13, IEEE
Li, M., D. Subhraveti, A.R. Butt, A. Khasymski and P. Xplore Press, Atlanta, GA, USA, pp: 1532-1537.
Sarkar, 2012. CAM: A topology aware minimum DOI: 10.1109/GLOCOM.2013.6831291
cost flow based resource manager for mapreduce Newsome, J. and D. Song, 2003. GEM: Graph
applications in the cloud. Proceedings of the 21st EMbedding for routing and data-centric storage in
International Symposium on High-Performance sensor networks without geographic information.
Parallel and Distributed Computing, Jun. 18-22, Proceedings of the 1st International Conference on
ACM, Delft, The Netherlands, pp: 211-222. Embedded Networked Sensor Systems, Nov. 05-07,
DOI: 10.1145/2287076.2287110 ACM, Los Angeles, California, USA, pp: 76-88.
Liu, D., X. Jia and I. Stojmenovic, 2007. Quorum and DOI: 10.1145/958491.958501
connected dominating sets based location service in Peng, Z., B. Xu, D. Cui, W. Lin and X.A. Wang, 2015.
wireless ad hoc, sensor and actuator networks. Deployment method of virtual machine cluster based
Comput. Commun., 30: 3627-3643. on energy minimization and graph cuts theory.
DOI: 10.1016/j.comcom.2007.07.001 Proceedings of the 10th International Conference on
Loutas, N., E. kamateri and E. Tarabanis, 2011. A P2P, Parallel, Grid, Cloud and Internet Computing,
semantic interoperability framework for cloud Nov. 4-6, IEEE Xplore Press, Krakow, Poland, pp:
platform as a service. Proceedings of the IEEE 3rd 800-803. DOI: 10.1109/3PGCIC.2015.51
International Conference on Cloud Computing Poola, D., K. Ramamohanarao and R. Buyya, 2014.
Technology and Science, Nov. 29-Dec. 1, IEEE Fault-tolerant workflow scheduling using spot
Xplore Press, Athens, Greece, pp: 280-287. instances on clouds. Proc. Comput. Sci., 29: 523-533.
DOI: 10.1109/CloudCom.2011.45 DOI: 10.1016/j.procs.2014.05.047

170
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy / Journal of Computer Science 2018, 14 (2): 163.172
DOI: 10.3844/jcssp.2018.163.172

Potluri, A. and A. Singh, 2012. A greedy heuristic and Tang, Y., Y. Zhang and H. Chen, 2008. A parallel
its variants for minimum capacitated dominating set. shortest path algorithm based on graph-partitioning
Proceedings of the International Conference on and iterative correcting. Proceedings of the 10th IEEE
Contemporary Computing, (CCC’ 12), Springer, International Conference on High Performance
Berlin, Heidelberg, pp: 28-39. Computing and Communications, Sept. 25-27, IEEE
DOI: 10.1007/978-3-642-32129-0_9 Xplorev Press, Dalian, China, pp: 155-161.
Prasanna, N.L., K. Sravanthi and N. Sudhakar, 2014. DOI: 10.1109/HPCC.2008.113
Applications of graph labeling in major areas of Tsai, L. and W. Liao, 2015. StarCube: An on-demand
computer science. Int. J. Res. Comput. Commun. and cost-effective framework for cloud data
Technol., 3: 819-823. center networks with performance guarantee.
IEEE Trans. Cloud Comput.
Radojević, B. and M. Žagar, 2011. Analysis of issues
DOI: 10.1109/TCC.2015.2464818
with load balancing algorithms in hosted (cloud)
Vecchiola, C., R.N. Calheiros, D. Karunamoorthy and R.
environments. Proceedings of the 34th International
Buyya, 2012. Deadline-driven provisioning of
Convention MIPRO, May 23-27, IEEE Xplore resources for scientific applications in hybrid clouds
Press, Opatija, Croatia, pp: 416-420. with Aneka. Future Generat. Comput. Syst., 28: 58-65.
Rimal, B.P., E. Choi and I. Lumb, 2009. A taxonomy DOI: 10.1016/j.future.2011.05.008
and survey of cloud computing systems. Venugopal, S. and R. Buyya, 2006. A set coverage-
Proceedings of the 5th International Joint conference based mapping heuristic for scheduling distributed
on INC, IMS and IDC, IEEE Xplore Press, Seoul, data-intensive applications on global grids.
South Korea, pp: 44-51. Proceedings of the 7th IEEE/ACM International
DOI: 10.1109/NCM.2009.218 Conference on Grid Computing, Sept. 28-29, IEEE
Sahni, J. and D.P. Vidyarthi, 2015. A cost-effective Xplore Press, Barcelona, Spain, pp: 238-245.
deadline-constrained dynamic scheduling algorithm DOI: 10.1109/ICGRID.2006.311021
for scientific workflows in a cloud environment. Verbelen, T., T. Stevens, F. De Turck and B. Dhoedt,
IEEE Trans. Cloud Comput. 2013. Graph partitioning algorithms for
DOI: 10.1109/TCC.2015.2451649 optimizing software deployment in mobile cloud
Scheideler, C., A.W. Richa and P. Santi, 2008. An O(log computing. Future Generat. Comput. Syst., 29:
n) dominating set protocol for wireless ad-hoc 451-459. DOI: 10.1016/j.future.2012.07.003
networks under the physical interference model. Vishwanath, K.V. and N. Nagappan, 2010.
Proceedings of the 9th ACM International Characterizing cloud computing hardware
Symposium on Mobile Ad Hoc Networking and reliability. Proceedings of the 1st ACM Symposium
Computing, May 26-30, ACM, Hong Kong, Hong on Cloud Computing, Jun. 10-11, ACM,
Kong, China, pp: 91-100. Indianapolis, Indiana, USA, pp: 193-204.
DOI: 10.1145/1374618.1374632 DOI: 10.1145/1807128.1807161
Shahzad, F., 2014. State-of-the-art survey on cloud Waibel, P., C. Hochreiner and S. Schulte, 2016. Cost-
computing security challenges, approaches and efficient data redundancy in the cloud. Proceedings
solutions. Proc. Comput. Sci., 37: 357-362. of the IEEE 9th International Conference on
DOI: 10.1016/j.procs.2014.08.053 Service-Oriented Computing and Applications, Nov.
Smith, D.M., D.W. Cearley and D.C. Plummer, 2009. 4-6, IEEE Xplore Press, Macau, China, pp: 1-9.
Key issues for cloud computing. Gartner. DOI: 10.1109/SOCA.2016.12
Subashini, S. and V. Kavitha, 2011. A survey on security Wang, W., Y. Wang, X.Y. Li, W.Z. Song and O. Frieder,
issues in service delivery models of cloud 2006. Efficient interference-aware TDMA link
computing. J. Netw. Comput. Applic., 34: 1-11. scheduling for static wireless networks. Proceedings
DOI: 10.1016/j.jnca.2010.07.006 of the 12th Annual International Conference on
Tamura, H. and K. Nakano, 2011. On applications of Mobile Computing and Networking, Sept. 23-29,
graph/network theory to problems in communication ACM, Los Angeles, CA, USA, pp: 262-273.
systems. ECTI Trans. Comput. Inform. Technol., 5: DOI: 10.1145/1161089.1161119
15-21. Wang, X., T. Tiropanis and H.C. Davis, 2012.
Tamura, H., M. Sengoku and S. Shinoda, 2008. Edge Evaluating graph traversal algorithms for distributed
coloring problem of graph theory considering SPARQL query optimization. Proceedings of the
interference on network coding. Proceedings of the Semantic Web: Joint International Semantic
23rd International Technical conference on Technology Conference, Dec. 4-7, Springer,
Circuits/Systems, Computers and Commnications, Hangzhou, China, pp: 210-225.
(SCC’ 08), pp: 301-304. DOI: 10.1007/978-3-642-29923-0_14

171
Kanniga Devi Rangaswamy and Murugaboopathi Gurusamy / Journal of Computer Science 2018, 14 (2): 163.172
DOI: 10.3844/jcssp.2018.163.172

Wang, Y. and X.Y. Li, 2006. Minimum power Xie, G., R. Li, X. Xiao and Y. Chen, 2014. A high-
assignment in wireless ad hoc networks with spanner performance DAG task scheduling algorithm for
property, J. Combinat. Optimiz., 11: 99-112. heterogeneous networked embedded systems.
DOI: 10.1023/A:1012319418150 Proceedings of the IEEE 28th International
Ward, J.S. and A. Barker, 2014. Observing the clouds: A Conference on Advanced Information Networking
and Applications, May 13-16, IEEE Xplore Press,
survey and taxonomy of cloud monitoring. J. Cloud
Victoria, BC, Canada, pp: 1011-1016.
Comput.: Adv. Syst. Applic., 3: 24-24. DOI: 10.1109/AINA.2014.123
DOI: 10.1186/s13677-014-0024-2 Xie, W., M. Goyal, H. Hosseini, J. Martocci and Y.
Warneke, D. and O. Kao, 2011. Exploiting dynamic Bashir et al., 2010. Routing loops in DAG-based
resource allocation for efficient parallel data processing low power and lossy networks. Proceedings of the
in the cloud. IEEE Trans. Parallel Distributed Syst., 22: 24th IEEE International Conference on Advanced
985-997. DOI: 10.1109/TPDS.2011.65 Information Networking and Applications, Apr. 20-23,
Wood, T., P. Shenoy, A. Venkataramani and M. Yousif, IEEE Xplore Press, Perth, WA, Australia, pp: 888-
2007. Black-box and gray-box strategies for virtual 895. DOI: 10.1109/AINA.2010.126
machine migration. Proceedings of the 4th USENIX Zegzhda, D.P. and A.V. Nikolsky, 2014. Formal security
Conference on Networked Systems Design and model for virtual machine hypervisors in cloud
Implementation, Apr. 11-13, USENIX Association, computing systems. Nonlinear Phenomena Complex
Cambridge, MA, pp: 229-242. Syst., 17: 253-262.
Wu, C. and R. Buyya, 2015. Cloud Data Centers and Zhang, Q., L. Cheng and R. Boutaba, 2010. Cloud
Cost Modeling: A Complete Guide to Planning, computing: State of-the-art and research challenges.
Designing and Building a Cloud Data Center. 1st J. Internet Services Applic., 1: 7-18.
Edn., Morgan Kaufmann, Waltham, MA, DOI: 10.1007/s13174-010-0007-6
ISBN-10: 012801413X, pp: 512. Zheng, Z., T.C. Zhou, M.R. Lyu and I. King, 2012.
Wu, J. and H. Li, 1999. On calculating connected Component ranking for fault-tolerant cloud
dominating set for efficient routing in ad hoc wireless applications. IEEE Trans. Services Comput., 5:
networks. Proceedings of the 3rd International 1097-1106. DOI: 10.1109/TSC.2011.42
Workshop on Discrete Algorithms and Methods for Zhou, Y., O. Grygorash and T.F. Hain, 2011.
Mobile Computing and Communications, Aug. 20- Clustering with minimum spanning trees. Int. J.
20, ACM, Seattle, Washington, USA, pp: 7-14. Artificial Intell. Tools, 20: 139-177.
DOI: 10.1142/S0218213011000061
DOI: 10.1145/313239.313261
Zissis, D. and D. Lekkas, 2012. Addressing cloud
Xiao, W. and B. Parhami, 2003. Some conclusions on computing security issues. Future Generat. Comput.
cayley digraphs and their applications to Syst., 28: 583-592. DOI: 10.1016/j.future.2010.12.006
interconnection networks. Proceedings of the 2nd
International Conference on Grid and Cooperative
Computing, Dec. 7-10, Springer, Shanghai, China,
pp: 408-412. DOI: 10.1007/978-3-540-24680-0_67

172