Exploiting implicit parallelism in functional programs with SLAM

DOI10.1007/3-540-45361-X_2MaRDI QIDQ2763375zbMATH OpenFDO

Authors John Sargeant, Chris Kirkham, Ian Watson

Publication date 15 January 2002

Published in Lecture Notes in Computer Science (Search for Journal in Brave)

Full work available at URL http://link.springer.de/link/service/series/0558/bibs/2011/20110019

Mathematics Subject Classification ID

Theory of programming languages (68N15) Functional programming and lambda calculus (68N18) Computing methodologies and applications (68U99)

Review text

An effective execution model is a vital component of any general-purpose implicitly parallel programming system. We introduce SLAM (Spreading Load with Active Messages), an execution model which overcomes many of the problems with previous approaches. SLAM is efficient enough to operate at low granularity without hardware support, and has other necessary properties. Compiling for SLAM presents an unusual set of problems, and we describe how this is done from UFO-Lite, a simplified version of the United Functions and Objects programming language. Linear speedups are obtained for a program with irregular, fine-grain, parallelism on stock hardware. For the entire collection see [Zbl 0977.68824].

Recommendations

• Algorithm + strategy = parallelism
• A statistically allocated parallel functional language
• Semi-explicit parallel programming in a purely functional style: GpH
• Implicitly threaded parallelism in Manticore
• Comparing parallel functional languages: Programming and performance

This page was built for publication: Exploiting implicit parallelism in functional programs with SLAM

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2763375)