{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/78642"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/78642","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Low-overhead scheduling for improving performance of scientific applications","abstract":"Application performance can degrade significantly due to node-local load imbalances during application execution on a large number of SMP nodes. These imbalances can arise from the machine, operating system, or the application itself. Although dynamic load balancing within a node can mitigate imbalances, such load balancing is challenging because of its impact to data movement and synchronization overhead. We developed a series of scheduling strategies that mitigate imbalances without incurring high overhead. Our strategies provide performance gains for various HPC codes, and perform better than widely known scheduling strategies such as OpenMP guided scheduling. Our developed scheme and methodology allows for scaling applications to next-generation clusters of SMPs with minimal application programmer intervention. We expect these techniques to be increasingly useful for future machines approaching exascale.","abstract_html":"Application performance can degrade significantly due to node-local load imbalances during application execution on a large number of SMP nodes. These imbalances can arise from the machine, operating system, or the application itself. Although dynamic load balancing within a node can mitigate imbalances, such load balancing is challenging because of its impact to data movement and synchronization overhead. We developed a series of scheduling strategies that mitigate imbalances without incurring high overhead. Our strategies provide performance gains for various HPC codes, and perform better than widely known scheduling strategies such as OpenMP guided scheduling. Our developed scheme and methodology allows for scaling applications to next-generation clusters of SMPs with minimal application programmer intervention. We expect these techniques to be increasingly useful for future machines approaching exascale.","abstract_has_math":false,"creators":["Kale, Vivek"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Gropp, William D.","de Supinski, Bronis","Garzaran, Maria J.","Padua, David A."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2015,"date_issued":"2015-07-22T22:33:40Z","date_published":"2015-07-22T22:33:40Z","updated_at":"2026-07-22T22:26:12Z","subjects":["multicore architecture","dynamic scheduling","scientific computing","performance tuning"],"languages":["en"],"rights":["Copyright 2015 Vivek Kale"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/78642","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Gropp, William D.","de Supinski, Bronis","Garzaran, Maria J.","Padua, David A."]},{"key":"dc:creator","label":"Author","values":["Kale, Vivek"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2015-07-22T22:33:40Z","2017-07-23T09:15:17Z","2015-05","2015-04-24","2015-5"]},{"key":"dc:type","label":"Dc Type","values":["text"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["multicore architecture","dynamic scheduling","scientific computing","performance tuning"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2015 Vivek Kale"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/78642"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Application performance can degrade significantly due to node-local load imbalances during application execution on a large number of SMP nodes. These imbalances can arise from the machine, operating system, or the application itself. Although dynamic load balancing within a node can mitigate imbalances, such load balancing is challenging because of its impact to data movement and synchronization overhead. We developed a series of scheduling strategies that mitigate imbalances without incurring high overhead. Our strategies provide performance gains for various HPC codes, and perform better than widely known scheduling strategies such as OpenMP guided scheduling. Our developed scheme and methodology allows for scaling applications to next-generation clusters of SMPs with minimal application programmer intervention. We expect these techniques to be increasingly useful for future machines approaching exascale.","Submission published under a 24 month embargo labeled 'U of I only', the embargo will last until 2017-05-01","The student, Vivek Kale, accepted the attached license on 2015-04-21 at 12:10.","The student, Vivek Kale, submitted this Dissertation for approval on 2015-04-21 at 12:22.","This Dissertation was approved for publication on 2015-04-24 at 10:48.","DSpace SAF Submission Ingestion Package generated from Vireo submission #7981 on 2015-07-22 at 14:18:24","Made available in DSpace on 2015-07-22T22:33:40Z (GMT). No. of bitstreams: 2 KALE-DISSERTATION-2015.pdf: 4985114 bytes, checksum: 93d5e68e9eabd103bc7abe9397313820 (MD5) LICENSE.txt: 4207 bytes, checksum: 34e53eef9f99e4a19f0c6e3a3690247b (MD5) Previous issue date: 2015-04-24","Embargo set by: Seth Robbins for item 79883 Lift date: 2017-07-22T22:34:16Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","U of I Only Restriction Lifted for Item 79883 on 2017-07-23T09:15:17Z."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Low-overhead scheduling for improving performance of scientific applications"]}]}],"canonical_facts":{"dc:contributor":["Gropp, William D.","de Supinski, Bronis","Garzaran, Maria J.","Padua, David A."],"dc:creator":["Kale, Vivek"],"dc:date":["2015-07-22T22:33:40Z","2017-07-23T09:15:17Z","2015-05","2015-04-24","2015-5"],"dc:description":["Application performance can degrade significantly due to node-local load imbalances during application execution on a large number of SMP nodes. These imbalances can arise from the machine, operating system, or the application itself. Although dynamic load balancing within a node can mitigate imbalances, such load balancing is challenging because of its impact to data movement and synchronization overhead. We developed a series of scheduling strategies that mitigate imbalances without incurring high overhead. Our strategies provide performance gains for various HPC codes, and perform better than widely known scheduling strategies such as OpenMP guided scheduling. Our developed scheme and methodology allows for scaling applications to next-generation clusters of SMPs with minimal application programmer intervention. We expect these techniques to be increasingly useful for future machines approaching exascale.","Submission published under a 24 month embargo labeled 'U of I only', the embargo will last until 2017-05-01","The student, Vivek Kale, accepted the attached license on 2015-04-21 at 12:10.","The student, Vivek Kale, submitted this Dissertation for approval on 2015-04-21 at 12:22.","This Dissertation was approved for publication on 2015-04-24 at 10:48.","DSpace SAF Submission Ingestion Package generated from Vireo submission #7981 on 2015-07-22 at 14:18:24","Made available in DSpace on 2015-07-22T22:33:40Z (GMT). No. of bitstreams: 2 KALE-DISSERTATION-2015.pdf: 4985114 bytes, checksum: 93d5e68e9eabd103bc7abe9397313820 (MD5) LICENSE.txt: 4207 bytes, checksum: 34e53eef9f99e4a19f0c6e3a3690247b (MD5) Previous issue date: 2015-04-24","Embargo set by: Seth Robbins for item 79883 Lift date: 2017-07-22T22:34:16Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","U of I Only Restriction Lifted for Item 79883 on 2017-07-23T09:15:17Z."],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/78642"],"dc:language":["en"],"dc:rights":["Copyright 2015 Vivek Kale"],"dc:subject":["multicore architecture","dynamic scheduling","scientific computing","performance tuning"],"dc:title":["Low-overhead scheduling for improving performance of scientific applications"],"dc:type":["text"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:26:12Z"}