DocumentCode :
627490
Title :
Power-effiicent resource allocation in MapReduce clusters
Author :
Kaiqi Xiong ; Yuxiong He
Author_Institution :
Coll. of Comput. & Inf. Sci, Rochester Inst. of Technol., Rochester, NY, USA
fYear :
2013
fDate :
27-31 May 2013
Firstpage :
603
Lastpage :
608
Abstract :
MapReduce has recently evolved in data-intensive parallel computing. It is a programming model for processing large data sets. The implementation of MapReduce typically runs on a large scale of cluster computing systems consisting of thousands of commodity machines. Such cluster computing systems are called MapReduce clusters. The high power consumption of MapReduce clusters has become a major concern since hundreds of MapReduce programs are implemented and thousands of MapReduce jobs are executed in such clusters like Amazon´s Elastic MapReduce Clusters every day. Power management becomes one of the most important problems in MapReduce clusters. Furthermore, the availability of MapReduce clusters plays an essential role in the delivery of quality of services (QoS) for customer services. In this paper, we investigate the problem of resource allocation for power management in MapReduce clusters. Specifically, we propose resource allocation approaches to minimizing the mean end-to-end delay of customer jobs or services under the constraints of the energy consumption and the availability of MapReduce clusters and to minimizing the energy consumption of MapReduce clusters under the availability of MapReduce clusters and the mean end-to-end delay of customer jobs or services. Numerical experiments demonstrate that the proposed approaches are applicable and efficient to solve these resource allocation problems for power management in MapReduce clusters.
Keywords :
customer services; parallel programming; power aware computing; quality of service; resource allocation; workstation clusters; MapReduce cluster computing systems; commodity machines; customer services; data-intensive parallel computing; end-to-end delay; energy consumption; large data set processing; power consumption; power management; power-effiicent resource allocation; programming model; quality of services; Availability; Clustering algorithms; Delays; Energy consumption; Quality of service; Resource management; Servers;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Integrated Network Management (IM 2013), 2013 IFIP/IEEE International Symposium on
Conference_Location :
Ghent
Print_ISBN :
978-1-4673-5229-1
Type :
conf
Filename :
6573039
Link To Document :
بازگشت