Workload-Adaptive Configuration Tuning for Hierarchical Cloud Schedulers

doi:10.1109/TPDS.2019.2923197

	Workload-Adaptive Configuration Tuning for Hierarchical Cloud Schedulers
	Han, Rui 1; Liu, Chi Harold 1; Zong, Zan 2; Chen, Lydia Y.3; Liu, Wending 1; Wang, Siyi 4; Zhan, Jianfeng 4
	2019-12-01
发表期刊	IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
ISSN	1045-9219
卷号	30 期号:12 页码:2879-2895
摘要	Cluster schedulers provide flexible resource sharing mechanism for best-effort cloud jobs, which occupy a majority in modern datacenters. Properly tuning a scheduler & x0027;s configurations is the key to these jobs' performance because it decides how to allocate resources among them. Today & x0027;s cloud scheduling systems usually rely on cluster operators to set the configuration and thus overlook the potential performance improvement through optimally configuring the scheduler according to the heterogeneous and dynamic cloud workloads. In this paper, we introduce AdaptiveConfig, a run-time configurator for cluster schedulers that automatically adapts to the changing workload and resource status in two steps. First, a comparison approach estimates jobs' performances under different configurations and diverse scheduling scenarios. The key idea here is to transform a scheduler & x0027;s resource allocation mechanism and their variable influence factors (configurations, scheduling constraints, available resources, and workload status) into business rules and facts in a rule engine, thereby reasoning about these correlated factors in job performance comparison. Second, a workload-adaptive optimizer transforms the cluster-level searching of huge configuration space into an equivalent dynamic programming problem that can be efficiently solved at scale. We implement AdaptiveConfig on the popular YARN Capacity and Fair schedulers and demonstrate its effectiveness using real-world Facebook and Google workloads, i.e., successfully finding best configurations for most of scheduling scenarios and considerably reducing latencies by a factor of two with low optimization time.
关键词	Resource management Scheduling Google Yarn Tuning Facebook Task analysis Cloud datacenter cluster scheduler configuration job latency YARN
DOI	10.1109/TPDS.2019.2923197
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Plan of China[2018YFB1003701] ; National Key Research and Development Plan of China[2018YFB1003700] ; National Natural Science Foundation of China[61872337]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS记录号	WOS:000498569400019
出版者	IEEE COMPUTER SOC
引用统计	被引频次：14[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/14957
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Liu, Chi Harold
作者单位	1.Beijing Inst Technol, Beijing Shi 100091, Peoples R China 2.Tsinghua Univ, Beijing Shi 100091, Peoples R China 3.Delft Univ Technol, NL-2628 CD Delft, Netherlands 4.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Han, Rui,Liu, Chi Harold,Zong, Zan,et al. Workload-Adaptive Configuration Tuning for Hierarchical Cloud Schedulers[J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,2019,30(12):2879-2895.
APA	Han, Rui.,Liu, Chi Harold.,Zong, Zan.,Chen, Lydia Y..,Liu, Wending.,...&Zhan, Jianfeng.(2019).Workload-Adaptive Configuration Tuning for Hierarchical Cloud Schedulers.IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,30(12),2879-2895.
MLA	Han, Rui,et al."Workload-Adaptive Configuration Tuning for Hierarchical Cloud Schedulers".IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 30.12(2019):2879-2895.