January 05, 2011
Gong, Gu & Wilkes Win Best Paper Award
Congratulations to Zhenhuan Gong, computer science PhD student, Dr. Xiaohui (Helen) Gu, assistant professor of computer science at NC State, and John Wilkes, a researcher at Google, whose co-authored paper entitled, “PRESS: Predictive Elastic ReSource Scaling for Cloud Systems”, was recently recognized with the Best Paper Award at the 6th IEEE International Conference on Network and Services Management (CNSM) in Niagra Falls, Canada in October 2010.
In this paper, the authors present a novel PRedictive Elastic ReSource Scaling (PRESS) scheme for cloud systems. Cloud systems require elastic resource allocation to minimize resource provisioning costs while meeting service level objectives (SLOs). PRESS unobtrusively extracts fine-grained dynamic patterns in application resource demands and adjusts their resource allocations automatically. This approach leverages light weight signal processing and statistical learning algorithms to achieve online predictions of dynamic application resource requirements. They implemented the PRESS system on Xen and tested it using RUBiS and an application load trace from Google. Experiments show that good resource prediction accuracy can be achieved with less than 5% over-estimation error and near zero under-estimation error, and elastic resource scaling can both significantly reduce resource waste and SLO violations.
The paper was selected for the award out of 176 submitted papers and 27 accepted papers. To read the award-winning paper, click here.
CNSM is a premier conference in the general area of network and services management. This conference was sponsored by IEEE, COMSOC and IFIP. For more information on the CNSM conference, click here.
Return To News Homepage