Optimizing Task Distribution and Shared-Variable Accesses in an Asynchronous-Partitioned-Global-Address-Space Programming Model Open Access
- Other title
- Type of item
- Degree grantor
University of Alberta
- Author or creator
Paudel, Jeeva S
- Supervisor and department
Amaral, Jose Nelson (Computing Science)
- Examining committee member and department
Szafron, Duane (Computing Science)
Pingali, Keshav (Computer Science)
Cockburn, Bruce (Electrical and Computer Engineering)
Lu, Paul (Computing Science)
Department of Computing Science
- Date accepted
- Graduation date
Doctor of Philosophy
- Degree level
High-performance programming systems employ a wide range of techniques to improve the performance of parallel and distributed applications on large scale machines. This dissertation identifies a novel opportunity of load balancing, proposes a new approach for workload distribution, and presents a profiling-based framework to automatically select coherence protocols aimed at specific patterns of shared-variable accesses. These approaches strike a balance between the tight budgets for run-time optimization and the exposition of new opportunities to improve the running time of applications. A prototype designed to evaluate these ideas is integrated into the X10 programming system. An empirical evaluation of these ideas, using large applications with diverse patterns of parallelism and communication, indicates that they can be applied widely and that they have significant performance merits.
- Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.
- Citation for previous publication
Jeeva Paudel and Jose Nelson Amaral. Using Cowichan Problems to Investigate the Programmability of X10 Programming System. In ACM SIGPLAN X10 Workshop, San Jose, CA, USA, 2011.Jeeva Paudel and Jose Nelson Amaral. Hybrid Parallel Task Placement in Irregular Applications. Journal of Parallel and Distributed Computing. doi:10.1016/j.jpdc.2014.09.014, 2014.Jeeva Paudel and Jose Nelson Amaral. Stratified Sampling for Even Workload Partitioning. In Proceedings of the 23rd International Conference on Parallel Architectures and Compilation Techniques. Edmonton, AB, Canada, 2014.Jeeva Paudel, Olivier Tardieu, and Jose Nelson Amaral. Hybrid Parallel Task Placement in X10. In ACM SIGPLAN X10 Workshop, Seattle, Washington, USA, 2013.Jeeva Paudel, Olivier Tardieu, and Jose Nelson Amaral. On the Merits of Distributed Work-Stealing on Selective Locality-Aware Tasks. In International Conference on Parallel Processing. Lyon, France, 2013.
- Date Uploaded
- Date Modified
- Audit Status
- Audits have not yet been run on this file.
File format: pdf (PDF/A)
Mime type: application/pdf
File size: 4003021
Last modified: 2015:10:21 22:25:51-06:00
Original checksum: 98b55c8af8a39e52febf9f80c536ad2b
Activity of users you follow