Optimizing Task Distribution and Shared-Variable Accesses in an Asynchronous-Partitioned-Global-Address-Space Programming Model

  • Author / Creator
    Paudel, Jeeva S
  • 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.

  • Subjects / Keywords
  • Graduation date
  • Type of Item
  • Degree
    Doctor of Philosophy
  • DOI
  • License
    This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.
  • Language
  • Institution
    University of Alberta
  • Degree level
  • Department
    • Department of Computing Science
  • Supervisor / co-supervisor and their department(s)
    • Amaral, Jose Nelson (Computing Science)
  • Examining committee members and their departments
    • Pingali, Keshav (Computer Science)
    • Szafron, Duane (Computing Science)
    • Cockburn, Bruce (Electrical and Computer Engineering)
    • Lu, Paul (Computing Science)