Paper Info
Title
Adaptive Fork-Heuristics For Software Thread-Level Speculation
Abstract
Fork-heuristics play a key role in software Thread-Level Speculation (TLS). Current fork-heuristics either lack real parallel execution environment information to accurately evaluate fork points and/or focus on hardware-TLS implementation which cannot be directly applied to software TLS. This paper proposes adaptive fork-heuristics as well as a feedback-based selection technique to overcome the problems. Adaptive fork-heuristics insert and speculate on all potential fork/join points and purely rely on the runtime system to disable inappropriate ones. Feedback-based selection produces parallelized programs with ideal speedups using log files generated by adaptive heuristics. Experiments of three scientific computing benchmarks on a 64-core machine show that feedback-based selection and adaptive heuristics achieve more than 88% and 50% speedups of the manual-parallel version, respectively. For the Barnes-Hut benchmark, feedback-based selection is 49% faster than the manual-parallel version.
Year
DOI
Venue
2013
10.1007/978-3-642-55224-3_49
PARALLEL PROCESSING AND APPLIED MATHEMATICS (PPAM 2013), PT I
Keywords
Field
DocType
Software thread-level speculation, Fork heuristics, Automatic parallelization, Performance tuning
Fork (system call),Speculation,Computer science,Parallel computing,Speculative multithreading,Software,Heuristics,Performance tuning,Runtime system,Automatic parallelization
Conference
Volume
ISSN
Citations 
8384
0302-9743
0
PageRank 
References 
Authors
0.34
11
2
Name
Order
Citations
PageRank
Zhen Cao100.34
Clark Verbrugge241139.15