Service Performance Analysis of Cloud Computing Systems
As cloud applications are evolving into missioncritical applications, evaluating their performance becomes an important task in the design and implementation of cloud systems. Because cloud systems deliver services, this work is mainly concerned with the analysis of cloud service performance, in particular, how fast the cloud can provide the requested service. Average response time is a common metric used to assess the quality of service requirements in the cloud computing service performance. However, this metric cannot address all the concerns of a customer; in many cloud applications, a customer may be more concerned with a percentile of response time that characterizes a threshold response time so that a certain percentage of requests can be served in under the threshold value. In this paper, we study a computing service performance model for the cloud infrastructure and assess the percentile of response time considering the impact of fault recovery using queueing theory. The relationships among the number of service resources, service rate, and percentile of response time are also studied through numerical examples.
Cloud computing Performance evaluation Percentile of response time
CHAONAN WANG JINGFANG HUANG ZHAOYANG ZHANG LIUDONG XING HONGGANG WANG
Department of Electrical and Computer Engineering, University of Massachusetts, Dartmouth, MA 02747, USA
国际会议
北京
英文
176-181
2011-06-20（万方平台首次上网日期，不代表论文的发表时间）