{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/18248"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/18248","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Load Balancing Regular Meshes on SMPS with MPI","abstract":"Domain decomposition for regular meshes on parallel computers has traditionally been performed by attempting to exactly partition the work among the available processors (now cores). However, these strategies often do not consider the inherent system noise which can hinder MPI application scalability to emerging peta-scale machines with 10000+ nodes. In this work, we suggest a solution that uses a tunable hybrid static/dynamic scheduling strategy that can be incorporated into current MPI implementations of mesh codes. By applying this strategy to a 3D jacobi algorithm, we achieve performance gains of at least 16% for 64 SMP nodes.","abstract_html":"Domain decomposition for regular meshes on parallel computers has traditionally been performed by attempting to exactly partition the work among the available processors (now cores). However, these strategies often do not consider the inherent system noise which can hinder MPI application scalability to emerging peta-scale machines with 10000+ nodes. In this work, we suggest a solution that uses a tunable hybrid static/dynamic scheduling strategy that can be incorporated into current MPI implementations of mesh codes. By applying this strategy to a 3D jacobi algorithm, we achieve performance gains of at least 16% for 64 SMP nodes.","abstract_has_math":false,"creators":["Kale, Vivek"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Gropp, William D."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2011,"date_issued":"2011-01-14T22:41:12Z","date_published":"2011-01-14T22:41:12Z","updated_at":"2026-07-22T22:25:09Z","subjects":["Regular Meshes","Scientific Computing","Message-Passing Interface (MPI)","Performance Tuning","Hybrid MPI-pthreads model"],"languages":["en"],"rights":["Copyright 2010 Vivek Kale"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/18248","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Gropp, William D."]},{"key":"dc:creator","label":"Author","values":["Kale, Vivek"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2011-01-14T22:41:12Z","2010-12"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"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":["Regular Meshes","Scientific Computing","Message-Passing Interface (MPI)","Performance Tuning","Hybrid MPI-pthreads model"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2010 Vivek Kale"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/18248"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Domain decomposition for regular meshes on parallel computers has traditionally been performed by attempting to exactly partition the work among the available processors (now cores). However, these strategies often do not consider the inherent system noise which can hinder MPI application scalability to emerging peta-scale machines with 10000+ nodes. In this work, we suggest a solution that uses a tunable hybrid static/dynamic scheduling strategy that can be incorporated into current MPI implementations of mesh codes. By applying this strategy to a 3D jacobi algorithm, we achieve performance gains of at least 16% for 64 SMP nodes.","Item withdrawn by Mark Zulauf (zulauf@illinois.edu) on 2010-12-08T22:21:22Z Item was in collections: University of Illinois Theses & Dissertations (ID: 1) No. of bitstreams: 2 thesis.tex: 2279 bytes, checksum: 2552522259b133a8dfe18c4410dd1761 (MD5) Kale_Vivek.pdf: 628609 bytes, checksum: 6f3e27f9e91e346c8f2263913d9478ca (MD5)","Made available in DSpace on 2011-01-14T22:41:12Z (GMT). No. of bitstreams: 4 Kale_Vivek.pdf: 628574 bytes, checksum: 422bac06fdbb0a2f19ebe6ebc4fb38a3 (MD5) license.txt: 4057 bytes, checksum: c74d52388c1ad0661f52c718727aaf3b (MD5) thesis.tex: 2279 bytes, checksum: 2552522259b133a8dfe18c4410dd1761 (MD5) 1_Kale_Vivek.pdf: 628574 bytes, checksum: 422bac06fdbb0a2f19ebe6ebc4fb38a3 (MD5)"]},{"key":"dc:title","label":"Title","values":["Load Balancing Regular Meshes on SMPS with MPI"]}]}],"canonical_facts":{"dc:contributor":["Gropp, William D."],"dc:creator":["Kale, Vivek"],"dc:date":["2011-01-14T22:41:12Z","2010-12"],"dc:description":["Domain decomposition for regular meshes on parallel computers has traditionally been performed by attempting to exactly partition the work among the available processors (now cores). However, these strategies often do not consider the inherent system noise which can hinder MPI application scalability to emerging peta-scale machines with 10000+ nodes. In this work, we suggest a solution that uses a tunable hybrid static/dynamic scheduling strategy that can be incorporated into current MPI implementations of mesh codes. By applying this strategy to a 3D jacobi algorithm, we achieve performance gains of at least 16% for 64 SMP nodes.","Item withdrawn by Mark Zulauf (zulauf@illinois.edu) on 2010-12-08T22:21:22Z Item was in collections: University of Illinois Theses & Dissertations (ID: 1) No. of bitstreams: 2 thesis.tex: 2279 bytes, checksum: 2552522259b133a8dfe18c4410dd1761 (MD5) Kale_Vivek.pdf: 628609 bytes, checksum: 6f3e27f9e91e346c8f2263913d9478ca (MD5)","Made available in DSpace on 2011-01-14T22:41:12Z (GMT). No. of bitstreams: 4 Kale_Vivek.pdf: 628574 bytes, checksum: 422bac06fdbb0a2f19ebe6ebc4fb38a3 (MD5) license.txt: 4057 bytes, checksum: c74d52388c1ad0661f52c718727aaf3b (MD5) thesis.tex: 2279 bytes, checksum: 2552522259b133a8dfe18c4410dd1761 (MD5) 1_Kale_Vivek.pdf: 628574 bytes, checksum: 422bac06fdbb0a2f19ebe6ebc4fb38a3 (MD5)"],"dc:identifier":["http://hdl.handle.net/2142/18248"],"dc:language":["en"],"dc:rights":["Copyright 2010 Vivek Kale"],"dc:subject":["Regular Meshes","Scientific Computing","Message-Passing Interface (MPI)","Performance Tuning","Hybrid MPI-pthreads model"],"dc:title":["Load Balancing Regular Meshes on SMPS with MPI"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:25:09Z"}