Static scheduling in clouds

Thomas A. Henzinger, Anmol V. Singh, Damien Zufferey, Vasu Singh, Thomas Wies

Research output: Contribution to conferencePaperpeer-review

Abstract

Cloud computing aims to give users virtually unlimited pay-per-use computing resources without the burden of managing the underlying infrastructure. We present a new job execution environment Flextic that exploits scalable static scheduling techniques to provide the user with a flexible pricing model, such as a tradeoff between different degrees of execution speed and execution price, and at the same time, reduce scheduling overhead for the cloud provider. We have evaluated a prototype of Flextic on Amazon EC2 and compared it against Hadoop. For various data parallel jobs from machine learning, image processing, and gene sequencing that we considered, Flextic has low scheduling overhead and reduces job duration by up to 15% compared to Hadoop, a dynamic cloud scheduler.

Original languageEnglish (US)
StatePublished - 2011
Event3rd USENIX Workshop on Hot Topics in Cloud Computing, HotCloud 2011 - Portland, United States
Duration: Jun 14 2011Jun 15 2011

Conference

Conference3rd USENIX Workshop on Hot Topics in Cloud Computing, HotCloud 2011
CountryUnited States
CityPortland
Period6/14/116/15/11

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Networks and Communications
  • Computer Science (miscellaneous)

Fingerprint Dive into the research topics of 'Static scheduling in clouds'. Together they form a unique fingerprint.

Cite this