Continuous adaptive object-code re-optimization framework

Howard Chen, Jiwei Lu, Wei Chung Hsu, Pen Chung Yew

Research output: Contribution to journalArticlepeer-review

16 Scopus citations

Abstract

ynamic optimization presents opportunities for finding run-time bottlenecks and deploying optimizations in statically compiled programs. In this paper, we discuss our current implementation of our hardware sampling based dynamic optimization framework and applying our dynamic optimization system to various SPEC2000 benchmarks compiled with the ORC compiler at optimization level O2 and executed on an Itanium-2 machine. We use our optimization system to apply memory prefetching optimizations, improving the performance of multiple benchmark programs.

Original languageEnglish (US)
Pages (from-to)241-255
Number of pages15
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3189
StatePublished - Dec 1 2004

Fingerprint

Dive into the research topics of 'Continuous adaptive object-code re-optimization framework'. Together they form a unique fingerprint.

Cite this