A power-aware scheduling of MapReduce applications in the cloud

Ying Li, Hongli Zhang, Kyong Hoon Kim

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

17 Scopus citations

Abstract

Cloud computing is an emerging computing technology for large data center that maintains computational resources through the internet, rather than on local computers. The large data centers maintain Cloud computing applications with lots of cost because of power consumption, which results in a new research issue, called Green Cloud computing. Since MapReduce is one of popular Cloud computing models, this paper focuses on how to reduce energy of MapReduce applications. Thus, we propose a new power-aware MapReduce application model to be used for power-aware computing with consideration of users' requirements. We also provide a scheduling algorithm for MapReduce applications in heterogeneous Cloud resources and suggest power-aware schemes in order to reduce the total energy. Throughout simulation results, we show that the proposed scheduling algorithm saves more energy than static schemes.

Original languageEnglish
Title of host publicationProceedings - IEEE 9th International Conference on Dependable, Autonomic and Secure Computing, DASC 2011
Pages613-620
Number of pages8
DOIs
StatePublished - 2011
Event9th IEEE Int. Conf. on Dependable, Autonomic and Secure Comput., DASC 2011, incl. 9th Int. Conf. on Pervasive Intelligence and Computing, PICom 2011, 9th Int. Symp. on Embedded Computing, EmbeddedCom 2011, 1st Int. Conf. on Cloud and Green Comput.CGC - Sydney, NSW, Australia
Duration: 12 Dec 201114 Dec 2011

Publication series

NameProceedings - IEEE 9th International Conference on Dependable, Autonomic and Secure Computing, DASC 2011

Conference

Conference9th IEEE Int. Conf. on Dependable, Autonomic and Secure Comput., DASC 2011, incl. 9th Int. Conf. on Pervasive Intelligence and Computing, PICom 2011, 9th Int. Symp. on Embedded Computing, EmbeddedCom 2011, 1st Int. Conf. on Cloud and Green Comput.CGC
Country/TerritoryAustralia
CitySydney, NSW
Period12/12/1114/12/11

Keywords

  • Cloud
  • MapReduce
  • power-aware
  • scheduling

Fingerprint

Dive into the research topics of 'A power-aware scheduling of MapReduce applications in the cloud'. Together they form a unique fingerprint.

Cite this