Towards unobtrusive VM live migration for cloud computing platforms

Akane Koto, Hiroshi Yamada, Kei Ohmura, Kenji Kono

Research output: Chapter in Book/Report/Conference proceedingConference contribution

35 Citations (Scopus)

Abstract

Live migration of virtual machines (VMs) is an attractive technique for managing cloud computing platforms such as those for load balancing and maintenance of physical machines. However, the execution of live migration significantly consumes computational resources on the source and destination, thus causing migration noise that degrades the performance of the VMs collocated on these hosts during live migration. Migration noise lowers the total throughput of data centers and make it difficult to carry out live migration to handle unpredictable increases in workload such as sudden request bursts and flash crowds. This paper introduces SonicMigration, which is an approach to shortening total migration time by pruning the VM memory to be transferred. Preliminary experiments with our prototype implemented on Xen 4.1.0 and Linux 2.6.38 show the migration time with our prototype is up to 68.3% shorter than that for Xen-based live migration and it successfully reduces network traffic by up to 83.9%.

Original languageEnglish
Title of host publicationProceedings of the Asia-Pacific Workshop on Systems, APSYS'12
DOIs
Publication statusPublished - 2012 Oct 8
EventAsia-Pacific Workshop on Systems, APSYS 2012 - Seoul, Korea, Republic of
Duration: 2012 Jul 232012 Jul 24

Publication series

NameProceedings of the Asia-Pacific Workshop on Systems, APSYS'12

Other

OtherAsia-Pacific Workshop on Systems, APSYS 2012
CountryKorea, Republic of
CitySeoul
Period12/7/2312/7/24

ASJC Scopus subject areas

  • Control and Systems Engineering

Fingerprint Dive into the research topics of 'Towards unobtrusive VM live migration for cloud computing platforms'. Together they form a unique fingerprint.

  • Cite this

    Koto, A., Yamada, H., Ohmura, K., & Kono, K. (2012). Towards unobtrusive VM live migration for cloud computing platforms. In Proceedings of the Asia-Pacific Workshop on Systems, APSYS'12 (Proceedings of the Asia-Pacific Workshop on Systems, APSYS'12). https://doi.org/10.1145/2349896.2349903