{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/89120"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/89120","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Optimizing parallel I/O performance of HPC applications","abstract":"Parallel I/O is an essential component of modern High Performance Computing (HPC). Obtaining good I/O performance for a broad range of applications on diverse HPC platforms is a major challenge, in part because of complex inter-dependencies between I/O middleware and hardware. The parallel file system and I/O middleware layers all offer optimization parameters that can, in theory, result in better I/O performance. Unfortunately, the right combination of parameters is highly dependent on the application, HPC platform, and problem size/concurrency. Scientific application developers do not have the time or expertise to take on the substantial burden of identifying good parameters for each problem configuration. They resort to using system defaults, a choice that frequently results in poor I/O performance. We expect this problem to be compounded on exascale class machines, which will likely have a deeper software stack with hierarchically arranged hardware resources. We present a line of solution to this problem containing an autotuning system for optimizing I/O performance, I/O performance modeling, I/O tuning, I/O kernel generation, and I/O patterns. We demonstrate the value of these solution across platforms, applications, and at scale.","abstract_html":"Parallel I/O is an essential component of modern High Performance Computing (HPC). Obtaining good I/O performance for a broad range of applications on diverse HPC platforms is a major challenge, in part because of complex inter-dependencies between I/O middleware and hardware. The parallel file system and I/O middleware layers all offer optimization parameters that can, in theory, result in better I/O performance. Unfortunately, the right combination of parameters is highly dependent on the application, HPC platform, and problem size/concurrency. Scientific application developers do not have the time or expertise to take on the substantial burden of identifying good parameters for each problem configuration. They resort to using system defaults, a choice that frequently results in poor I/O performance. We expect this problem to be compounded on exascale class machines, which will likely have a deeper software stack with hierarchically arranged hardware resources. We present a line of solution to this problem containing an autotuning system for optimizing I/O performance, I/O performance modeling, I/O tuning, I/O kernel generation, and I/O patterns. We demonstrate the value of these solution across platforms, applications, and at scale.","abstract_has_math":false,"creators":["Behzad, Babak"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Snir, Marc","Winslett, Marianne","Gropp, William","Hildebrand, Dean"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2016,"date_issued":"2016-03-02T20:23:36Z","date_published":"2016-03-02T20:23:36Z","updated_at":"2026-07-22T22:26:32Z","subjects":["High Performance Computing (HPC)","Parallel Computing","Parallel I/O","Big Data","Storage Performance Tuning","Autotuning"],"languages":["en"],"rights":["Copyright 2015 Babak Behzad"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/89120","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Snir, Marc","Winslett, Marianne","Gropp, William","Hildebrand, Dean"]},{"key":"dc:creator","label":"Author","values":["Behzad, Babak"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2016-03-02T20:23:36Z","2018-03-03T10:15:34Z","2015-11-23","2015-12"]},{"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":["High Performance Computing (HPC)","Parallel Computing","Parallel I/O","Big Data","Storage Performance Tuning","Autotuning"]}]},{"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 Babak Behzad"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/89120"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Parallel I/O is an essential component of modern High Performance Computing (HPC). Obtaining good I/O performance for a broad range of applications on diverse HPC platforms is a major challenge, in part because of complex inter-dependencies between I/O middleware and hardware. The parallel file system and I/O middleware layers all offer optimization parameters that can, in theory, result in better I/O performance. Unfortunately, the right combination of parameters is highly dependent on the application, HPC platform, and problem size/concurrency. Scientific application developers do not have the time or expertise to take on the substantial burden of identifying good parameters for each problem configuration. They resort to using system defaults, a choice that frequently results in poor I/O performance. We expect this problem to be compounded on exascale class machines, which will likely have a deeper software stack with hierarchically arranged hardware resources. We present a line of solution to this problem containing an autotuning system for optimizing I/O performance, I/O performance modeling, I/O tuning, I/O kernel generation, and I/O patterns. We demonstrate the value of these solution across platforms, applications, and at scale.","Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2017-12-01","The student, Babak Behzad, accepted the attached license on 2015-11-20 at 13:31.","The student, Babak Behzad, submitted this Dissertation for approval on 2015-11-20 at 13:45.","This Dissertation was approved for publication on 2015-11-23 at 13:41.","DSpace SAF Submission Ingestion Package generated from Vireo submission #8816 on 2016-03-02 at 14:06:01","Made available in DSpace on 2016-03-02T20:23:36Z (GMT). No. of bitstreams: 4 BEHZAD-DISSERTATION-2015.pdf: 2579773 bytes, checksum: e393f126dc8ca1229a99a983f6997d0b (MD5) ProQuestSign-BabakBehzad.pdf: 268035 bytes, checksum: f29cad9dd3b5d70629f7fa2b72875efc (MD5) BabakBehzad_CS_PhDThesis_Final_2015.zip: 19558300 bytes, checksum: ab9daeaa06c6144e16b86542046ac95e (MD5) LICENSE.txt: 4209 bytes, checksum: 648e218a7b53159b5f1fbf5c97c5c231 (MD5) Previous issue date: 2015-11-23","Embargo set by: Seth Robbins for item 91322 Lift date: 2018-03-02T20:24:31Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","U of I Only Restriction Lifted for Item 91322 on 2018-03-03T10:15:34Z."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Optimizing parallel I/O performance of HPC applications"]}]}],"canonical_facts":{"dc:contributor":["Snir, Marc","Winslett, Marianne","Gropp, William","Hildebrand, Dean"],"dc:creator":["Behzad, Babak"],"dc:date":["2016-03-02T20:23:36Z","2018-03-03T10:15:34Z","2015-11-23","2015-12"],"dc:description":["Parallel I/O is an essential component of modern High Performance Computing (HPC). Obtaining good I/O performance for a broad range of applications on diverse HPC platforms is a major challenge, in part because of complex inter-dependencies between I/O middleware and hardware. The parallel file system and I/O middleware layers all offer optimization parameters that can, in theory, result in better I/O performance. Unfortunately, the right combination of parameters is highly dependent on the application, HPC platform, and problem size/concurrency. Scientific application developers do not have the time or expertise to take on the substantial burden of identifying good parameters for each problem configuration. They resort to using system defaults, a choice that frequently results in poor I/O performance. We expect this problem to be compounded on exascale class machines, which will likely have a deeper software stack with hierarchically arranged hardware resources. We present a line of solution to this problem containing an autotuning system for optimizing I/O performance, I/O performance modeling, I/O tuning, I/O kernel generation, and I/O patterns. We demonstrate the value of these solution across platforms, applications, and at scale.","Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2017-12-01","The student, Babak Behzad, accepted the attached license on 2015-11-20 at 13:31.","The student, Babak Behzad, submitted this Dissertation for approval on 2015-11-20 at 13:45.","This Dissertation was approved for publication on 2015-11-23 at 13:41.","DSpace SAF Submission Ingestion Package generated from Vireo submission #8816 on 2016-03-02 at 14:06:01","Made available in DSpace on 2016-03-02T20:23:36Z (GMT). No. of bitstreams: 4 BEHZAD-DISSERTATION-2015.pdf: 2579773 bytes, checksum: e393f126dc8ca1229a99a983f6997d0b (MD5) ProQuestSign-BabakBehzad.pdf: 268035 bytes, checksum: f29cad9dd3b5d70629f7fa2b72875efc (MD5) BabakBehzad_CS_PhDThesis_Final_2015.zip: 19558300 bytes, checksum: ab9daeaa06c6144e16b86542046ac95e (MD5) LICENSE.txt: 4209 bytes, checksum: 648e218a7b53159b5f1fbf5c97c5c231 (MD5) Previous issue date: 2015-11-23","Embargo set by: Seth Robbins for item 91322 Lift date: 2018-03-02T20:24:31Z Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD system","U of I Only Restriction Lifted for Item 91322 on 2018-03-03T10:15:34Z."],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/89120"],"dc:language":["en"],"dc:rights":["Copyright 2015 Babak Behzad"],"dc:subject":["High Performance Computing (HPC)","Parallel Computing","Parallel I/O","Big Data","Storage Performance Tuning","Autotuning"],"dc:title":["Optimizing parallel I/O performance of HPC 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":"2026-07-22T22:26:32Z"}