System Level Benchmarking of Public Clouds

Author:

Ahuja Sanjay P.1

Affiliation:

1. University of North Florida, United States

Abstract

It is important for cloud users to be able to evaluate and compare different cloud services to achieve high performance and maximize cost savings. To that end, this research benchmarked Amazon web services elastic compute cloud and Rackspace cloud infrastructure and compared the results for the two public cloud providers. The intent of the study was to determine how these selected providers perform with regards to system parameter usage and hence three system-level benchmarks: STREAM, IOR, and NPB-EP were run on different configurations to provide an insight to the cloud users in selection of provider on VM clusters of 1,2,4,6, and 8 nodes. The clusters were created with similar virtual machines from both providers. The benchmarks examined bandwidth, I/O and CPU performance. A comparison of results for the two providers is presented graphically and T-test applied to determine if differences are significant. Observations were taken at multiple times at different time periods on weekdays and weekends to examine variance of cloud performance.

Publisher

IGI Global

Subject

Computer Networks and Communications,Computer Science Applications,Human-Computer Interaction

Reference12 articles.

1. A Multi-Agent-Based Data Collection and Aggregation Model for Fog-Enabled Cloud Monitoring

2. Cabello, R. D. (2017). Rackspace Open Cloud reference architecture. https://docs.rackspace.com/support/how-to/Rackspace-open-cloud-reference-architecture/?_ga=1.114351619.549215010.1487796249

3. Filtering of Distributed Denial of Services (DDoS) Attacks in Cloud Computing Environment

4. Price Comparison for Infrastructure as a Service.;S. E.Kihal;European Conference on Information Systems,2012

5. Leitner, P. and Cito, J. (2016). Patterns in the Chaos—A Study of Performance Variation and Predictability in Public IaaS Clouds. ACM Transactions on Internet Technology, 16, pp. 1.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3