Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Distributed Array Data Management on NUMA Multiprocessors Tarek S. Abdelrahman and Thomas N. Wong
 

Summary: Distributed Array Data Management on NUMA Multiprocessors
Tarek S. Abdelrahman and Thomas N. Wong
Department of Electrical and Computer Engineering
University of Toronto
Toronto, Ontario, M5S 1A4
Canada
Abstract
Management of program data to reduce false shar-
ing and improve locality is critical for scaling perfor-
mance on NUMA multiprocessors. We use HPF-like
directives to partition and place arrays in data-parallel
applications on Hector, a shared-memory NUMA mul-
tiprocessor. We present experimental results that
demonstrate the magnitude of the performance im-
provement attainable when our proposed array man-
agement schemes are used instead of the operating sys-
tem management policies. We then describe a com-
piler system for automating the partitioning and place-
ment of arrays. The compiler uses a number of tech-
niques that exploit Hector's shared memory architec-

  

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

 

Collections: Computer Technologies and Information Sciences