TitlePower Consumption Modeling and Prediction in a Hybrid CPU-GPU-MIC Supercomputer
Publication TypeConference Paper
Year of Publication2016
AuthorsSirbu, A, Babaoglu, O
Conference NameEuropean Conference on Parallel Processing
Date Published08/2016
PublisherSpringer International Publishing
Conference LocationGrenoble, France
ISBN Number978-3-319-43659-3
KeywordsHigh performance computing, Hybrid system, Job power modeling, Job power prediction, Support vector regression
Abstract

Power consumption is a major obstacle for High Performance Computing (HPC) systems in their quest towards the holy grail of ExaFLOP performance. Significant advances in power efficiency have to be made before this goal can be attained and accurate modeling is an essential step towards power efficiency by optimizing system operating parameters to match dynamic energy needs. In this paper we present a study of power consumption by jobs in Eurora, a hybrid CPU-GPU-MIC system installed at the largest Italian data center. Using data from a dedicated monitoring framework, we build a data-driven model of power consumption for each user in the system and use it to predict the power requirements of future jobs. We are able to achieve good prediction results for over 80 % of the users in the system. For the remaining users, we identify possible reasons why prediction performance is not as good. Possible applications for our predictive modeling results include scheduling optimization, power-aware billing and system-scale power modeling. All the scripts used for the study have been made available on GitHub.

URLhttp://link.springer.com/chapter/10.1007/978-3-319-43659-3_9
DOI10.1007/978-3-319-43659-3_9
Refereed DesignationRefereed
Research Line: