Adaptive Off-Line Tuning for Optimized Composition of Components for Heterogeneous Many-Core Systems

Lu Li, Usman Dastgeer, Christoph Kessler

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

Abstract / Description of output

In recent years heterogeneous multi-core systems have been given much attention. However, performance optimization on these platforms remains a big challenge. Optimizations performed by compilers are often limited due to lack of dynamic information and run time environment, which makes applications often not performance portable. One current approach is to provide multiple implementations for the same interface that could be used interchangeably depending on the call context, and expose the composition choices to a compiler, deployment-time composition tool and/or run-time system. Using off-line machine-learning techniques allows to improve the precision and reduce the run-time overhead of run-time composition and leads to an improvement of performance portability. In this work we extend the run-time composition mechanism in the PEPPHER composition tool by off-line composition and present an adaptive machine learning algorithm for generating compact and efficient dispatch data structures with low training time. As dispatch data structure we propose an adaptive decision tree structure, which implies an adaptive training algorithm that allows to control the trade-off between training time, dispatch precision and run-time dispatch overhead.
Original languageEnglish
Title of host publicationHigh Performance Computing for Computational Science - VECPAR 2012
EditorsMichel Daydé, Osni Marques, Kengo Nakajima
Place of PublicationBerlin, Heidelberg
PublisherSpringer Berlin Heidelberg
Pages329-345
Number of pages17
ISBN (Electronic)978-3-642-38718-0
ISBN (Print)978-3-642-38717-3
DOIs
Publication statusPublished - 2013
Event10th International Meeting on High-Performance Computing for Computational Science - Kobe, Japan
Duration: 17 Jul 201220 Jul 2012
http://nkl.cc.u-tokyo.ac.jp/VECPAR2012/

Publication series

NameLecture Notes in Computer Science
Volume7851

Conference

Conference10th International Meeting on High-Performance Computing for Computational Science
Abbreviated titleVECPAR 2012
Country/TerritoryJapan
CityKobe
Period17/07/1220/07/12
Internet address

Fingerprint

Dive into the research topics of 'Adaptive Off-Line Tuning for Optimized Composition of Components for Heterogeneous Many-Core Systems'. Together they form a unique fingerprint.

Cite this