2026-09-19T07:45:51Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/18323 2023-07-10 col_2142_5131 col_2142_10761 com_2142_5130 com_2142_10755 com_2142_234

Kale, Laxmikant V. Kale, Laxmikant V. Heath, Michael T. Zilles, Craig Jefferson, David Dooley, Isaac J. 2011-01-14T22:46:12Z 2011-01-14T22:46:12Z 2011-01-14T22:46:12Z The tuning of parallel programs on large distributed-memory machines today is usually a costly, and often extensive, manual process. Automatic tuning techniques can help reduce this manual burden. This dissertation investigates the utility of a new class of automatic tuning methods for large-scale parallel programs whereby each program exposes information about its behavior to the runtime system. This behavioral information enables a tuning framework to quickly find appropriate ways to reconfigure or steer the application towards better performance. This dissertation describes both new automatic tuning mechanisms within a parallel runtime system, and a new framework that automatically reconfigures the behavior or structure of the program through one or more control points. Control points are a novel type of tunable parameter provided by an application wherein it exposes tunable knobs and information about the behavioral effects expected to occur as each knob is varied in each direction. This behavioral information associated with each control point allows tuning algorithms to identify the direction in which a control point should be adjusted to fix observed performance problems. Multiple application case studies show that control points are useful mechanisms for dynamically reconfiguring applications to improve their performance. In these case studies, individual control points are examined to investigate how they can adjust diverse application behaviors including computational grain sizes, the amount of work offloaded to accelerators, the mapping of tasks to processors, the frequency of load balancing, and a communication throttling parameter. Item withdrawn by Mark Zulauf (zulauf@illinois.edu) on 2010-10-22T13:38:24Z Item was in collections: University of Illinois Theses & Dissertations (ID: 1) No. of bitstreams: 1 Dooley_Isaac.pdf: 5673417 bytes, checksum: de919bc140c5bc796354bb8d101e4577 (MD5) Made available in DSpace on 2011-01-14T22:46:12Z (GMT). No. of bitstreams: 2 Dooley_Isaac.pdf: 5673417 bytes, checksum: de919bc140c5bc796354bb8d101e4577 (MD5) license.txt: 4062 bytes, checksum: 0c3c55f636c8bec9376177fd698971fa (MD5) http://hdl.handle.net/2142/18323 en Copyright 2010 Isaac Dooley. Portions copyright 2010 IEEE. Reprinted, with permission, from Proceedings of 17th Annual International Conference on High Performance Computing (HiPC), and 12th Workshop on Advances in Parallel and Distributed Computing Models (APDCM) This material is posted here with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of the products or services of the University of Illinois at Urbana Champaign. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to pubs-permissions@ieee.org.By choosing to view this material, you agree to all provisions of the copyright laws protecting it. Parallel Programming Auto-tuning Automatic Performance Tuning Control Points Performance Tuning Distributed-memory parallel programming Intelligent runtime tuning of parallel applications with control points 2010-12 Computer Science 1434 Computer Science 0112 University of Illinois at Urbana-Champaign Dissertation Ph.D. PHD:Computer Science -UIUC 10KS0112PHD