울산대학교 | IT융합전공
본문바로가기
ender
대학원안내
대학원공지사항

대학원공지사항

[세미나 안내] Performance Tuning in Clouds - 3월 9일(금)
작성자 학**** 작성일 2012-03-08 조회수 513
일시: 3월 9일 금요일, 오전 11시 - 12시
장소: 7-505호실
강사: C. R. Das, Professor
소속: Dept. of ECE, The Pennsylvania State University
강연 제목: Performance Tuning in Clouds - A Multidimensional Approach
요약:

Cloud computing is an emerging paradigm that has the potential to transform the future computing landscape for many application domains.  With cloud computing, both hardware and software resources can be leased from cloud providers, thereby avoiding the capital investment on procuring and maintaining such infrastructures.  Therefore, various pay-as-you-go models like SaaS, PaaS, and IaaS are becoming attractive options to achieve better economy of scale, elasticity and availability.  However, a major roadblock for the wide-scale adoption of clouds is the performance unpredictability of applications due to factors such as system scale, workload dynamism, virtualization overheads, and resource sharing and contention.

In this talk, I will discuss several issues that are critical for performance enhancement in clouds and summarize some of our current efforts for alleviating the performance problem. In this context, two complementary scheduling techniques for enhancing cloud performance will be discussed. The first technique will highlight the importance of an important cloud workload property, called task placement constraints, and the second technique will summarize the impact of a fine-grained resource scheduling mechanism for data analytic systems like MapReduce.  The talk will conclude with a simple performance model, called D-factor, for estimating application performance.