Practical Iterative Optimization for the Data Center

doi:10.1145/2739048

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	Practical Iterative Optimization for the Data Center
	Fang, Shuangde ; Xu, Wenwen 2; Chen, Yang 1,2; Eeckhout, Lieven 3; Temam, Olivier ; Chen, Yunji ; Wu, Chengyong 2; Feng, Xiaobing 2
刊名	ACM TRANSACTIONS ON ARCHITECTURE AND CODE OPTIMIZATION
	2015-07-01
卷号	12 期号:2 页码:26
关键词	Design Performance Iterative optimization compiler MapReduce server data center co-run
ISSN号	1544-3566
DOI	10.1145/2739048
英文摘要	Iterative optimization is a simple but powerful approach that searches the best possible combination of compiler optimizations for a given workload. However, iterative optimization is plagued by several practical issues that prevent it from being widely used in practice: a large number of runs are required to find the best combination, the optimum combination is dataset dependent, and the exploration process incurs significant overhead that needs to be compensated for by performance benefits. Therefore, although iterative optimization has been shown to have a significant performance potential, it seldom is used in production compilers. In this article, we propose iterative optimization for the data center (IODC): we show that the data center offers a context in which all of the preceding hurdles can be overcome. The basic idea is to spawn different combinations across workers and recollect performance statistics at the master, which then evolves to the optimum combination of compiler optimizations. IODC carefully manages costs and benefits, and it is transparent to the end user. To bring IODC to practice, we evaluate it in the presence of co-runners to better reflect real-life data center operation with multiple applications co-running per server. We enhance IODC with the capability to find compatible co-runners along with a mechanism to dynamically adjust the level of aggressiveness to improve its robustness in the presence of co-running applications. We evaluate IODC using both Map Reduce and compute-intensive throughput server applications. To reflect the large number of users interacting with the system, we gather a very large collection of datasets (up to hundreds of millions of unique datasets per program), for a total storage of 16.4TB and 850 days of CPU time. We report an average performance improvement of 1.48x and up to 2.08x for five MapReduce applications, and 1.12x and up to 1.39x for nine server applications. Furthermore, our experiments demonstrate that IODC is effective in the presence of co-runners, improving performance by greater than 13% compared to the worst possible co-runner schedule.
资助项目	European Research Council under the European Community's Seventh Framework Programme / ERC[259295] ; Google Faculty Research Award ; Intel Collaborative Research Institute for Computational Intelligence (ICRI-CI) ; China 1000-talents program ; National Natural Science Foundation of China (NSFC)[61100163] ; National Natural Science Foundation of China (NSFC)[61133004] ; National Natural Science Foundation of China (NSFC)[61222204] ; National Natural Science Foundation of China (NSFC)[61221062] ; National Natural Science Foundation of China (NSFC)[61303158] ; National Natural Science Foundation of China (NSFC)[61432016] ; National Natural Science Foundation of China (NSFC)[61472396] ; National Natural Science Foundation of China (NSFC)[61473275] ; National High Technology Research and Development Program of China[2012AA012202] ; National High Technology Research and Development Program of China[2012AA010902] ; Strategic Priority Research Program of CAS[XDA06010403] ; International Collaboration Key Program of CAS[171111KYSB20130002] ; China 10000-talents program ; NSFC[60873057] ; NSFC[60921002] ; NSFC[60925009] ; NSFC[61033009] ; NSFC[61202055] ; NSFC[61402445] ; National Basic Research Program of China[2011CB302504]
WOS研究方向	Computer Science
语种	英语
出版者	ASSOC COMPUTING MACHINERY
WOS记录号	WOS:000357952000007
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/9589]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Fang, Shuangde
作者单位	1.Microsoft Res, Beijing, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 3.Univ Ghent, Dept Elect & Informat Syst, B-9000 Ghent, Belgium
推荐引用方式 GB/T 7714	Fang, Shuangde,Xu, Wenwen,Chen, Yang,et al. Practical Iterative Optimization for the Data Center[J]. ACM TRANSACTIONS ON ARCHITECTURE AND CODE OPTIMIZATION,2015,12(2):26.
APA	Fang, Shuangde.,Xu, Wenwen.,Chen, Yang.,Eeckhout, Lieven.,Temam, Olivier.,...&Feng, Xiaobing.(2015).Practical Iterative Optimization for the Data Center.ACM TRANSACTIONS ON ARCHITECTURE AND CODE OPTIMIZATION,12(2),26.
MLA	Fang, Shuangde,et al."Practical Iterative Optimization for the Data Center".ACM TRANSACTIONS ON ARCHITECTURE AND CODE OPTIMIZATION 12.2(2015):26.