{"title":"Performance and Power Consumption Measurement of Java Application Servers","authors":"Hitoshi Oi, Sho Niboshi","doi":"10.1109/MASCOTS.2012.68","DOIUrl":null,"url":null,"abstract":"In this paper, we present our in-progress project of modeling performance and power consumption of Java application servers using SPECjEnterprise2010. We run the workload on two application server using two different CPUs, AMD Phenom II and Intel Atom, and investigate performance and power consumption behaviors against the increasing system sizes. We have observed that: (1) CPU utilization draws non-linear functions of the system size and their shapes are different on Phenom and Atom. However, power consumption on both servers increase proportionally. (2) Browse transaction is the source of non-linearly in the CPU utilization. (3) Estimation of the CPU utilization from that of each transaction measured separately incurs large errors (up to 65%), while the errors in the estimation of the power consumption are relatively small (up to 4%).","PeriodicalId":278764,"journal":{"name":"2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"44 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 IEEE 20th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MASCOTS.2012.68","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
In this paper, we present our in-progress project of modeling performance and power consumption of Java application servers using SPECjEnterprise2010. We run the workload on two application server using two different CPUs, AMD Phenom II and Intel Atom, and investigate performance and power consumption behaviors against the increasing system sizes. We have observed that: (1) CPU utilization draws non-linear functions of the system size and their shapes are different on Phenom and Atom. However, power consumption on both servers increase proportionally. (2) Browse transaction is the source of non-linearly in the CPU utilization. (3) Estimation of the CPU utilization from that of each transaction measured separately incurs large errors (up to 65%), while the errors in the estimation of the power consumption are relatively small (up to 4%).