Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
To appear in Proceedings of ISCA24 (June, 1997) The Interaction of Software Prefetching with
 

Summary: To appear in Proceedings of ISCA­24 (June, 1997)
The Interaction of Software Prefetching with
ILP Processors in Shared­Memory Systems \Lambda
Parthasarathy Ranganathan, Vijay S. Pai, Hazim Abdel­Shafi, Sarita V. Adve
Department of Electrical and Computer Engineering
Rice University
Houston, Texas 77005
fparthas---vijaypai---shafi---saritag@rice.edu
Abstract
Current microprocessors aggressively exploit instruction­
level parallelism (ILP) through techniques such as multiple
issue, dynamic scheduling, and non­blocking reads. Recent
work has shown that memory latency remains a significant
performance bottleneck for shared­memory multiprocessor
systems built of such processors.
This paper provides the first study of the effectiveness
of software­controlled non­binding prefetching in shared­
memory multiprocessors built of state­of­the­art ILP­based
processors. We find that software prefetching results in sig­
nificant reductions in execution time (12% to 31%) for three

  

Source: Adve, Sarita - Department of Computer Science, University of Illinois at Urbana-Champaign

 

Collections: Computer Technologies and Information Sciences