On predicting the impact of resource redistributions in streaming applications

Merijn Verstraaten, Sven Bodo Scholz

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

Abstract

We propose a method for black-box performance modelling of executions of data-parallel array operations on shared memory multicore systems. Black-box performance modelling refers to the idea that the source code as well as its attendant compilation process are completely independent from the modelling itself. The performance model exclusively builds on observable behaviour available when executing the compiled code. From given input characteristics and previous runtime observations we predict overall runtimes in relation to the number of cores that can be exclusively used for the task. We show that using our technique our model's runtime predictions fall within 10% of the observed runtime. The paper describes the rationale as well as the technical details of the approach.We discuss several design choices of the technique and we experimentally explore their implications. We also discuss an online implementation of the proposed approach and we show that the model can be used very effectively in a streaming context.

Original languageEnglish
Title of host publicationProceedings of the 2014 ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming
PublisherAssociation for Computing Machinery
Pages76-81
Number of pages6
ISBN (Print)9781450329378
DOIs
Publication statusPublished - 2014
Event1st ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming - Edinburgh, United Kingdom
Duration: 12 Jun 201413 Jun 2014

Conference

Conference1st ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming
Abbreviated titleARRAY 2014 - Part of PLDI 2014
CountryUnited Kingdom
CityEdinburgh
Period12/06/1413/06/14

ASJC Scopus subject areas

  • Software

Fingerprint Dive into the research topics of 'On predicting the impact of resource redistributions in streaming applications'. Together they form a unique fingerprint.

Cite this