Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Impact of virtualization on cloud computing energy consumption: empirical study
Atiewi S., Abuhussein A., Saleh M.  ISCSIC 2018 (Proceedings of the 2nd International Symposium on Computer Science and Intelligent Control, Stockholm, Sweden, Sep 21-23, 2018)1-7.2018.Type:Proceedings
Date Reviewed: Apr 5 2019

Intensive computation in tasks such as weather forecasting and scientific solar system exploration requires high-speed processors and huge data storage. Very large data centers are likely to consume more energy as the number of servers substantially increases. Perhaps cloud computing technology that supports virtual machines (VMs) and virtualization practices, such as infrastructure as a service (IaaS), might reduce energy consumption. Atiewi et al. investigate the impact of two task scheduling algorithms on energy efficiency utilization in virtualized settings. They install and configure an extension of the NS2 network simulator, called the GreenCloud simulator, and then use it to construct a data center and a cloud network, link the data centers to the VMs, create processing requests and requirements of cloud users on the VMs, and monitor and compute metrics that impact energy consumption on cloud computers.

To execute the GreenCloud simulation, the input consists of the number of data centers, the number and specifications of VMs in the data center, the number and specifications of tasks for the data center, and either the green scheduler or the power saver scheduling algorithm (PSSA). The output parameters of the simulator include: the maximum task completion time, the executed number and failure number of user tasks, the data center and server loads, and the energy consumed by the data center. The data center load is the proportion of the data center size allocated to inbound tasks; each server load is its current load relative to the supreme computing capability. The data center energy consumption is the total energy consumed by the servers and switches.

How effective is the proposed PSSA for saving energy? The authors claim that, compared to the non-virtualized servers, the virtualized servers consumed less energy and improved the data center loads. Unfortunately, the simulation experimental results do not show any observable significant differences in the data center loads and the total energy consumed by the virtualized and non-virtualized servers. The data center with virtualized servers only outperformed the non-virtualized servers in reducing the energy required to complete tasks.

The authors clearly acknowledge the failure of the VMs to reduce the energy throughput. Is there something wrong with these simulation experiments that fail to focus on power consumed by network switches, that is, merging workloads and combining traffic? Cloud computing technology, energy, virtualization, and simulation experts should examine this question and recommend alternative solutions.

Reviewer:  Amos Olagunju Review #: CR146515 (1906-0233)
Bookmark and Share
  Reviewer Selected
Featured Reviewer
 
 
Performance of Systems (C.4 )
 
 
Cloud Computing (C.2.4 ... )
 
Would you recommend this review?
yes
no
Other reviews under "Performance of Systems": Date
A computer and communications network performance analysis primer
Stuck B., Arthurs E., Prentice-Hall, Inc., Upper Saddle River, NJ, 1985. Type: Book (9789780131639812)
Jun 1 1985
A mean value performance model for locking in databases
Tay Y., Suri R. (ed), Goodman N. Journal of the ACM 32(3): 618-651, 1985. Type: Article
Mar 1 1986
The relationship between benchmark tests and microcomputer price
Sircar S., Dave D. Communications of the ACM 29(3): 212-217, 1986. Type: Article
Nov 1 1986
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy