Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Exploiting Wavefront Parallelism on Large-Scale Shared-Memory Multiprocessors
 

Summary: Exploiting Wavefront Parallelism on
Large-Scale Shared-Memory Multiprocessors
Naraig Manjikian, Member, IEEE, and Tarek S. Abdelrahman, Member, IEEE
Abstract–Wavefront parallelism, in which parallelism is limited to hyperplanes in an iteration space, can arise when compilers apply
tiling to loop nests to enhance locality. Previous approaches for scheduling wavefront parallelism focused on maximizing parallelism,
balancing workloads, and reducing synchronization. In this paper, we show that on large-scale shared-memory multiprocessors,
locality is a crucial factor. We make the distinction between intratile and intertile locality and show that as the number of processors
grows, intertile locality becomes more important. We consider and experimentally evaluate existing strategies for scheduling wavefront
parallelism. We show that dynamic self-scheduling can be efficiently used on a small number of processors, but performs poorly at
large scale because it does not enhance intertile locality. By contrast, static scheduling strategies enhance intertile locality for small
tiles, maintaining parallelism and resulting in better performance at large scale. Results from a Convex SPP1000 multiprocessor
demonstrate the importance of taking intertile locality into account. Static scheduling outperforms dynamic self-scheduling by a factor
of up to 2.3 on 30 processors.
Index Terms–High-performance compilers, wavefront parallelism, cache locality, locality-enhancing loop transformations, tiling,
large-scale shared-memory multiprocessors.
ś
1 INTRODUCTION
LARGE-SCALE shared-memory multiprocessors are increas-
ingly used as platforms for high-performance
computing [6]. Several commercial and research sys-

  

Source: Abdelrahman, Tarek S. - Department of Electrical and Computer Engineering, University of Toronto

 

Collections: Computer Technologies and Information Sciences