A SURVEY ON REDUCING ENERGY SPRAWL IN CLOUD COMPUTING

Advanced Computational Intelligence: An International Journal (ACII), Vol.2, No.3, July 2015
DOI:10.5121/acii.2015.2303 23
A SURVEY ON REDUCING ENERGY SPRAWL IN
CLOUD COMPUTING
Malathi.P
M.E.Scholar, Department of Computer Science & Engineering, Nandha Engineering
College, Erode, Tamil Nadu, India
ABSTRACT
Cloud computing is the cluster of autonomic computing, grid computing and utility computing. Cloud
providers are there to rescue their customers from the problem of dynamism. The providers focus on
resource sharing and in improving the performance. Energy consumption is the major factor to degrade the
performance. Reducing energy sprawl will bloom the performance. This paper delineates the different
techniques involved in scheduling the workload of the servers in order to minimize the energy sprawl.
KEYWORDS
Cloud computing, Energy consumption, Virtualization, Task consolidation, Virtual machine
1. INTRODUCTION
Cloud computing is a cluster of various computing and it provides the package of computer
resources as a metered service. Cloud computing is an autonomic computing. The cloud providers
handle the needs of their clients such as dynamism, abstraction, resource sharing. The cloud
stacks are software as a service, platform as a service and infrastructure as a service.
Infrastructure as a service is to provide virtualization. Platform as a service is to use the platform
on web. Software as a service is directly consumed by end user. Cloud providers are there to
rescue their customers from the problem of dynamism. The providers focus on resource sharing
and in improving the performance. Energy consumption is the major factor to degrade the
performance.
75
150
275
0
100
200
300
2000 2005 2010
billion kwh/year
Figure 1.worldwide power consumption in data centers [7].

24
As shown in figure 1, the data centers have been consuming the energy raised up to 56% from
2005 to 2010 [7]. As koomey [23] spotted that, energy consumption in data centers always in hike
unless energy efficient techniques are developed and applied.to address the problem of energy
sprawl, to eliminate the wastage and inefficient usage of electricity in the computing resources.
The additional spark for the problem is the idle servers which consume up to 70% of their peak
power [24].
In this survey paper, energy harnessing techniques are discussed. The remaining paper is
organized as follows: Section 2 presents overview, Section 3 presents the literature review of the
existing methods of energy efficient techniques, and Section 4 presents conclusions.
2. OVERVIEW
2.1 Energy Consumption Problem
The major problem in cloud is energy sprawl. This is due to two reasons. One is manual faults
like improper scheduling and work overload [5, 10] and other is due to hike in monetary cost of
electricity [13]. The manual faults can be resolved by using task consolidation and virtualization.
Second problem can be resolved by harnessing the renewable energy [9].The overall power
consumption of the server [26] shown in figure 2.
memory
CPU quad core
fan
motherboard
NIC
PSU efficiency loss
PCIslots
Figure 2.power consumption of the server [26].
2.1.1 Task consolidation
The task consolidation is to enhance resource utilization in a while reduce energy sprawl by
assigning a set of tasks to a set of resources [21]. Task consolidation abases the amount of virtual
machines, labours and energy. It aims to consolidate a set of tasks to reduce overload and also
saves energy. The resource usage associated with the tasks is directly related to time constraints
[11].

25
2.1.2 Virtualization
Virtualization technology provides the flexible resource provisioning and migration of machine
state [5]. Due to hot spot, excess space capacity and load imbalance which migrates the machine.
Virtualization enables the consolidation, load balancing and hot spot mitigation [4]. It allocate
data center resources dynamically based on application demands and support green computing by
optimizing number of servers in use. The reasons for virtualization techniques are shown in
figure 3.
Figure 3.Reason for migrating machine.
3. EXISTING ENERGY EFFICIENT MODELS IN THE CLOUD
3.1. Probabilistic Consolidation of Virtual Machines
Jianying Luo et al. [16] shows that the virtual machine consolidation switch off the unloaded
server and migrates the overloaded machine to reduce energy and server sprawl. This approach
handle the multidimensional problem consolidated with respect to two resources CPU and RAM.
It was consolidated by using two probabilistic procedures assignment and migration. Ad- hoc
simulator is used for simulation. Assignment procedure is used for analytical study and
experimented by both assignment and migration procedures. It limits the energy usage and
balancing CPU bound.
3.2. Hint Based Execution of Workloads
Konstantino et al. [11] has proposed a hint based execution of workloads with Nefeli in cloud to
use the energy in efficient manner. Nefeli is a virtual infrastructure gateway. Hints are deployed
to collect all the information about the virtual machine. As per the hint they reschedule the virtual
machine. Nefeli a high level virtual machine placement policies and it adds a layer between the
user and infrastructure. During operation, Nefeli has obtained the following information such as
physical node property, physical infrastructure property, current status of each virtual machine
and virtual machine properties. This information is passed as a hint. The evaluation shows that the
performance and energy saving is increased.

26
3.3. Dynamic Consolidation of Virtual Machines
Fahimeh Farahnakian et al. [19] have proposed ant colony based virtual machine consolidation
which uses artificial ants to consolidate virtual machine into a reduced number of physical
machines based on the current resource requests and ants work in parallel to build virtual machine
migration. It composed with global and local agent. Global agent consolidates virtual machine
into reduced number of physical machine. Local agent detects physical machine status. It was
implemented by cloudsim toolkit. Result shows that this technique reduces the energy
consumption up to 53.4% than the dynamic virtual machine consolidation.
3.4. Virtual Machine Scheduling
Dong Jiankang et al. [20] leveraged virtual machine scheduling to solve the combination of bin
packing problem and quadratic assignment problem. It employs the two stage heuristic algorithm
with virtual machine placement and migration. Virtual machine placement is to meet the physical
capacity and network bandwidth. Virtual machine migration to minimize the migration costs to
optimize the network maximum link utilization to reduce the energy consumption. c++ is used to
develop the algorithm. Compared to random algorithm the energy consumption is low.
3.5. A Hierarchical Approach for the Resource Management
Bernadette Addis et al. [1] have developed a mixed-integer nonlinear optimization of resource
management based hierarchical framework. It uses the resource allocation policies to allocate the
resources. Local search algorithm is deployed. It was implemented by java (sun java 1.6). The
result shows that hierarchal framework reduces the energy sprawl.
3.6. Tabu Search Algorithm
Federico Larumbe et al. [2] leveraged tabu search algorithm to achieve the high quality of service,
low cost and low co2 emissions. It used tabu list to avoid the repeated solutions. Data centre get
the power from the nearest renewable resource and grid. The tabu search algorithm shows that
they find near optimal solution in short execution time and less energy consumption of 34.6
KWH than greedy approach.
3.7. Dynamic Heterogeneity-Aware Resource Provisioning
Carlo Mastroianni et al. [3] have proposed heterogeneity –aware capacity provisioning scheme
which use the k-means of clustering algorithm to divide the workload into distinct classes with
similar characteristics of resources and dynamically adjusting the virtual machines to minimize
the total energy consumption and scheduling delay. CPU utilization and memory utilization are
the parameters for evaluation. They implement this technique in matlab.
3.8. Virtual Machine Migrations
Mayank Mishra et al. [5] have proposed virtualization technology. Virtualization enables the
consolidation, load balancing and hot spot mitigation. Virtualization technology provides the
flexible resource provisioning and migration of machine state. Dynamic provisioning using
virtual machine migration follows two steps. A first step is to deploy the virtual machine. Second
step is to keep the resource monitoring engine which tracks the resource usage and performance.

27
3.9. Energy-aware Migration Algorithm
Mohammad H.AL Shyaeji et al. [25] have proposed an energy aware migration algorithm to save
energy in which they migrates the virtual machines. An algorithm composed of three parts such
as: Victim Selection, Target Server Selection, and Switch on Server. Victim selection is to switch
off the under loaded machines. Target server selection is to select the machine which and where
to be migrated during overloaded. Switch on server is to on the machine in sleep mode during
busy periods.
Table 1. Comparison of the existing methods
Techniques and Algorithms Hardware / Datasets Tools Parameter Analysis
Scalable distributed hierarchical
framework.
Intel Nehalem dual Socket
quad-core CPU @2.4 GHZ
with 24 GB of RAM
Running Ubuntu Linux
2011.4
IBM data
center.
Energy cost
Tabu search algorithm Hadoop cluster C++ Number of nodes
Energy cost
Heterogeneity-aware capacity
provisioning scheme
Heterogeneous cluster Matlab CPU utilization
Memory utilization
Virtualization technology. Not mentioned Not
mentioned
Not mentioned
Energy Aware Migration
Algorithm
No specific environmental
set up.
Custom
Built
Simulator
Load Balancing,
minimization of active
server
Energy aware task
consolidation(ETC)
Data center data sets Cloud sim
toolkit
Number of nodes
Workloads
Consolidation of virtual machine Data center data sets Ad-hoc
simulator
Number of active
servers,
Migrations per hour,
power
Hint Based Execution of
Workloads
Data center data sets Nefeli Energy consumption,
Time consumption
Dynamic consolidation of virtual
machines
Hadoop testbed Cloudsim
toolkit
Energy consumption ,
Number of virtual
machine migration
Virtual machine scheduling
algorithm
.
No specific environmental
set up.
C++ Energy consumption,
Total communication
Traffic,
Maximum link
utilization

28
3.10. Energy-Aware Task Consolidation
Ching-Hsien Hsu et al. [11] have developed a technique that reduces energy consumption is
energy-aware task consolidation (ETC) technique. ETC restricts CPU use below a specified peak
threshold. Task consolidation is the major work of ETC. When a task migrates to other virtual
clusters considered as network latency by energy cost model. Compared ETC with MAXUTIL
for evaluation. MAXUTIL is a greedy algorithm that aspires to maximize cloud computing
resources. The simulation result shows that 17% improvement over MAXUTIL.
4. CONCLUSIONS
The various techniques were discussed for reducing the energy sprawl in cloud computing. The
two amplifying factors are virtualization and task consolidation plays a vital role in optimizing
energy consumption. Techniques like harnessing renewable energy, frequency scaling and
workload consolidation enhance the performance of cloud by optimizing energy consumption.
These techniques are used to resolve energy consumption problem. Thus, this survey can be used
to enhance the energy consumption models by designing energy prediction models, energy
optimization models and energy consumption monitors for the cloud system.
REFERENCES
[1] Bernadette Addis, DaniloArdagna, Barbara Panicucci, Mark S. Squillante and Li Zhang, “A
Hierarchical Approach for the Resource Management of Very Large Cloud Platforms,” IEEE
Transactions on Dependable and Secure Computing, vol. 10, no. 5, September/October 2013.
[2] Federico Larumbe and Brunilde Sanso,” A Tabu Search Algorithm For The Location Of Data Centers
And Software Components In Green Cloud Computing Networks”, IEEE Transactions On Cloud
Computing, Vol. 1, No. 1, January-June 2013.
[3] Carlo Mastroianni, MichelaMeo and Giuseppe Papuzzo,”Dynamic Heterogeneity-Aware Resource
Provisioning In The Cloud”, IEEE Transactions On Cloud Computing, Vol. 1, No. 2, July-December
2013.
[4] Zhen Xiao, Weijia Song, and Qi Chen, ” Dynamic Resource Allocation Using Virtual Machines for
Cloud Computing Environment,” IEEE Transactions on Parallel and Distributed Systems, vol. 24, no.
6, June 2013.
[5] Mayank Mishra, Anwesha Das, Purushottam Kulkarni, and Anirudha Sahoo,” Dynamic Resource
Management Using Virtual Machine Migrations,” IEEE Communications Magazine, 0163-6804/12,
September 2012.
[6] Li Hongyou, Wang Jiangyong, Peng Jia, Wang Junfeng, Liu Tang, ”Energy-Aware Scheduling
Scheme Using Workload-Aware Consolidation Technique In Cloud Data Centres”, China
Communications , December 2014.
[7] ——, “Growth in data center electricity use 2005 to 2010,” Analytics Press, Tech. Rep., 2011.
[8] Michael Cardosa, Aameek Singh, Himabindu Pucha and Abhishek Chandra,” Exploiting Spatio-
Temporal Trade-offs For Energy-Aware Map reduce In The Cloud”, IEEE Transactions On
Computers, Vol. 61, No. 12, December 2012.
[9] UttamMandal, M.FarhanHabib, Shuqiang Zhang and Biswanath Mukherjee, Davis Massimo
Tornatore, Davis and Politecnico Di Milano” Greening the Cloud Using Renewable-Energy-Aware
Service Migration”, IEEE Network, November/December 2013.
[10] Wei Deng, Fangming Liu and Hai Jin,” Harnessing Renewable Energy In cloud Datacentres
Opportunities and Challenges”, IEEE Network, January/February 2014.
[11] Konstantinos Tsakalozos, Mema Roussopoulos, and Alex Delis,” Hint-Based Execution of Workloads
In Clouds With Nefeli”, IEEE Transactions On Parallel And Distributed Systems, Vol. 24, No. 7, July
2013.
[12] Ching-Hsien Hsu, KennD.Slagter, Shih-Chang Chen, Yeh –Chinh Chung,” Optimizing Energy
Consumption with Task Consolidation in Cloud”, Information Sciences, No. 3, March 2014.
[13] Chonglin Gu, Hejiao Huang, and Xiaohua Jia,” Power Metering For Virtual Machine In Cloud
Computing—Challenges And Opportunities”, IEEE Access, Vol. 2, Sep 2014.

29
[14] Jianguo Yao, Xue Liu, and Chen Zhang,” Predictive Electricity Cost Minimization Through Energy
Buffering In Data Centers”, IEEE Transactions On Smart grid, Vol.5,No.1,January2014.
[15] Carlo Mastroianni, MichelaMeo, and Giuseppe Papuzzo,” Probabilistic Consolidation Of Virtual
Machines In Self-Organizing Cloud Data Centers”, IEEE Transactions On Cloud Computing, Vol. 1,
No. 2, July-December 2013.
[16] Xiaomin Zhu, Laurence T. Yang, Huangke Chen, Ji Wang, Shu Yin and Xiaocheng Liu,” Real-Time
Tasks Oriented Energy-Aware Scheduling In Virtualized Clouds”, IEEE Transactions On Cloud
Computing, Vol. 2/April-June 2014.
[17] JianyingLuo, Lei Rao, and Xue Li,” Temporal Load Balancing With Service Delay Guarantees For
Data Center Energy Cost Optimization”, IEEE Transactions On Parallel And Distributed Systems,
Vol. 25, March 2014.
[18] Mehiar Dabbagh, Bechir Hamdaoui, Mohsen Guizani, and Ammar Rayes,” Toward Energy-Efficient
Cloud Computing Prediction, Consolidation, And Over commitment”, IEEE Network, March/April
2015.
[19] Weiwen Zhang, Yonggang Wen, and Hsiao-Hwa Chen,” Toward Transcoding as a Service Energy-
Efficient Offloading Policy for Green Mobile Cloud”, IEEE Network, November/December 2014.
[20] Fahimeh Farahnakian, Adnan Ashraf, Tapio Pahikkala, Pasi Liljeberg, Juha Plosila, Ivan Porres, And
Hannu Tenhunen,”Using Ant Colony System To Consolidate Vms For Green Cloud Computing”,
IEEE Transactions On Services Computing, Vol. 8, No. 2, March/April 2015.
[21] Dong Jiankang, Wang Hongbo, Li Yang yang, Cheng Shiduan,”Virtual Machine Scheduling For
Improving Energy Efficiency In Iaas Cloud”, China Communications , March 2014.
[22] Elizabeth Sylvester Mkoba, Mokhtar Abdullah Abdo Saif,” A Survey On Energy Efficient With Task
Consolidation In The Virtualized Cloud Computing Environment “, IJRET: International Journal of
Research in Engineering and Technology.
[23] J. G. Koomey, “Estimating total power consumption by servers in the US and the world,” Lawrence
Berkeley National Laboratory, Tech. Rep., 2007.
[24] X.Fan,W.D.Weber,andL.A.Barroso,“Power provisioning for a ware house-sized computer,” in
Proceedings of the 34th Annual International Symposium on Computer Architecture (ISCA), 2007,
pp. 13–23.
[25] Mohammad H.AL Shyaeji and M.D.Samrajesh, “An energy-aware virtual machine algorithm”,
International Conference on Advances in Computing and Communication(ICACC), pp. 242-246,
Aug. 2012.
[26] L. Minas and B. Ellison, Energy Efﬁciency for Information Technology: How to Reduce Power
Consumption in Servers and Data Centers. Intel Press, 2009.

A SURVEY ON REDUCING ENERGY SPRAWL IN CLOUD COMPUTING

More Related Content

What's hot (12)

Viewers also liked (15)

Similar to A SURVEY ON REDUCING ENERGY SPRAWL IN CLOUD COMPUTING (20)

Recently uploaded (20)

A SURVEY ON REDUCING ENERGY SPRAWL IN CLOUD COMPUTING