Skip to main content

Automatically exploiting cross-invocation parallelism using runtime information

Author(s): Huang, Jialu; Jablin, Thomas B; Beard, Stephen R; Johnson, Nick P; August, David I

To refer to this page use:
Abstract: Automatic parallelization is a promising approach to producing scalable multi-threaded programs for multicore architectures. Many existing automatic techniques only parallelize iterations within a loop invocation and synchronize threads at the end of each loop invocation. When parallel code contains many loop invocations, synchronization can easily become a performance bottleneck. Some automatic techniques address this problem by exploiting cross-invocation parallelism. These techniques use static analysis to partition iterations among threads to avoid crossthread dependences. However, this partitioning is not always achievable at compile-time, because program input determines dependence patterns at run-time. By contrast, this paper proposes DOMORE, the first automatic parallelization technique that uses runtime information to exploit additional cross-invocation parallelism. Instead of partitioning iterations statically, DOMORE dynamically detects crossthread dependences and synchronizes only when necessary. DOMORE consists of a compiler and a runtime library. At compile time, DOMORE automatically parallelizes loops and inserts a custom runtime engine into programs. At run-time, the engine observes dependences and synchronizes iterations only when necessary. For six programs, DOMORE achieves a geomean loop speedup of 2.1× over parallel execution without cross-invocation parallelization and of 3.2 × over sequential execution on eight cores.
Publication Date: 2013
Citation: Huang, Jialu, Thomas B. Jablin, Stephen R. Beard, Nick P. Johnson, and David I. August. "Automatically exploiting cross-invocation parallelism using runtime information." Proceedings of the 2013 IEEE/ACM International Symposium on Code Generation and Optimization (CGO) (2013): pp. 1-11. doi:10.1109/CGO.2013.6495001
DOI: 10.1109/CGO.2013.6495001
ISSN: 2164-2397
Pages: 1 - 11
Type of Material: Conference Article
Journal/Proceeding Title: Proceedings of the 2013 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)
Version: Author's manuscript

Items in OAR@Princeton are protected by copyright, with all rights reserved, unless otherwise indicated.