CFP last date
20 June 2024
Reseach Article

Disk I/O Performance-per-Watt Analysis for Cloud Computing

by Joseph Issa, Abdallah Kassem
International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
Volume 97 - Number 3
Year of Publication: 2014
Authors: Joseph Issa, Abdallah Kassem

Joseph Issa, Abdallah Kassem . Disk I/O Performance-per-Watt Analysis for Cloud Computing. International Journal of Computer Applications. 97, 3 ( July 2014), 23-27. DOI=10.5120/16988-7090

@article{ 10.5120/16988-7090,
author = { Joseph Issa, Abdallah Kassem },
title = { Disk I/O Performance-per-Watt Analysis for Cloud Computing },
journal = { International Journal of Computer Applications },
issue_date = { July 2014 },
volume = { 97 },
number = { 3 },
month = { July },
year = { 2014 },
issn = { 0975-8887 },
pages = { 23-27 },
numpages = {9},
url = { },
doi = { 10.5120/16988-7090 },
publisher = {Foundation of Computer Science (FCS), NY, USA},
address = {New York, USA}
%0 Journal Article
%1 2024-02-06T22:23:09.342632+05:30
%A Joseph Issa
%A Abdallah Kassem
%T Disk I/O Performance-per-Watt Analysis for Cloud Computing
%J International Journal of Computer Applications
%@ 0975-8887
%V 97
%N 3
%P 23-27
%D 2014
%I Foundation of Computer Science (FCS), NY, USA

Assessing disk I/O performance and power consumption associated with a given cloud workload is important especially for workloads that are bounded by disk I/O. Disk performance becomes a bottleneck for achieving higher performance and lower power consumption especially when memory size is not enough to process large blocks of data. This will lead to a negative impact on the Quality-of-Service (QoS). In this paper, we analyze disk I/O performance by assessing the disk bandwidth and latency for different reads and writes configurations for sequential and random patterns. The systems used are based on ATOM D525 and Xeon X5660 processors. We analyze power consumption for both systems and provide a performance-per-watt optimum operation point. We also propose an estimation method which estimates disk latency at different disk queue depth settings. The estimation method is verified to estimate disk latency with < 5% error margin.

  1. Hadoop: http://hadoop. apache. org
  2. D. R. Jiang, B. Ooi B. , L. Shi, and S. Wu. "The performance of mapreduce: an in-depth study". Proceedings of the VLDB Endowment
  3. G. Wang, A. Butt, P. Pandey, and K. Gupta, "Using realistic simulation for performance analysis of MapReduce setups". LSAP. ACM, 2009.
  4. T. Chao, H. Zhou, Y. He, and L. Zha, "A Dynamic MapReduce Scheduler for Heterogeneous Workloads". IEEE Computer Society, 2009.
  5. J. Issa and S. Figueira, "Hadoop and Memcached: Performance and Power Characterization and Analysis". Springer Journal of Cloud Computing:Advances, systems and Applications, 2012
  6. James Hendricks, et al. , "Improving small file performance in object-based storage", Technical report, Carnegie Mellon University Parallel Data Lab, 2006
  7. A. Manda et al, "Provisioning and Evaluating Multi-domain Networked Clouds for Hadoop-based Applications", Third International Conference on Cloud Computing Technology and Science (CloudCom), 2011
  8. J. Shafer, S. Rixner, and A. L. Cox, A. L. , "The Hadoop distributed filesystem: Balancing portability and performance", IEEE International Symposium on Performance Analysis of Systems &amp; Software (ISPASS), 2010
  9. Disk IO benchmarking in the cloud : http://blog. cloudharmony. com/2010/06/disk-io-benchmarking-in-cloud. html
  10. FIO: http://freecode. com/projects/fio
  11. Kenneth Hoste, Lieven Eeckhout, Hendrick Blockeel "Analyzing Commercial Processor Performance Numbers for Predicting Performance of Application on Interest
  12. S. Krishnaprasad , "Uses and abuses of Amdahl's law" December 2001 Journal of Computing Sciences in colleges" Volume 17, Issue 2
  13. Rafael H. Saavedra, Alan J. Smith. "Analysis of benchmark characteristics and benchmark performance prediction" Volume 14, Issue 4, (November 1996) ISSN:0734-20
Index Terms

Computer Science
Information Sciences


Disk I/O performance performance-per-watt analysis cloud computing