{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/90472"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/90472","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Optimization of communication intensive applications on HPC networks","abstract":"Communication is a necessary but overhead inducing component of parallel programming. Its impact on application design and performance is due to several related aspects of a parallel job execution: network topology, routing protocol, suitability of algorithm being used to the network, job placement, etc. This thesis is aimed at developing an understanding of how communication plays out on networks of high performance computing systems and exploring methods that can be used to improve communication performance of large scale applications. Broadly speaking, three topics have been studied in detail in this thesis. The first of these topics is task mapping and job placement on practical installations of torus and dragonfly networks. Next, use of supervised learning algorithms for conducting diagnostic studies of how communication evolves on networks is explored. Finally, efficacy of packet-level simulations for prediction-based studies of communication performance on different networks using different network parameters is analyzed. The primary contribution of this thesis is development of scalable diagnostic and prediction methods that can assist in the process of network designing, adapting applications to future systems, and optimizing execution of applications on existing systems. These meth- ods include a supervised learning approach, a functional modeling tool (called Damselfly), and a PDES-based packet level simulator (called TraceR), all of which are described in this thesis.","abstract_html":"Communication is a necessary but overhead inducing component of parallel programming. Its impact on application design and performance is due to several related aspects of a parallel job execution: network topology, routing protocol, suitability of algorithm being used to the network, job placement, etc. This thesis is aimed at developing an understanding of how communication plays out on networks of high performance computing systems and exploring methods that can be used to improve communication performance of large scale applications. Broadly speaking, three topics have been studied in detail in this thesis. The first of these topics is task mapping and job placement on practical installations of torus and dragonfly networks. Next, use of supervised learning algorithms for conducting diagnostic studies of how communication evolves on networks is explored. Finally, efficacy of packet-level simulations for prediction-based studies of communication performance on different networks using different network parameters is analyzed. The primary contribution of this thesis is development of scalable diagnostic and prediction methods that can assist in the process of network designing, adapting applications to future systems, and optimizing execution of applications on existing systems. These meth- ods include a supervised learning approach, a functional modeling tool (called Damselfly), and a PDES-based packet level simulator (called TraceR), all of which are described in this thesis.","abstract_has_math":false,"creators":["Jain, Nikhil"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Kalé, Laxmikant V","Gropp, William D.","Torrellas, Josep","Panda, Dhabaleswar K."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2016,"date_issued":"2016-07-07T19:52:47Z","date_published":"2016-07-07T19:52:47Z","updated_at":"2026-07-22T22:26:32Z","subjects":["Network","Communication","Parallel computing","Applications","Simulation","Modeling","Prediction","Machine learning","Topology-aware mapping"],"languages":["en"],"rights":["Copyright 2016, Nikhil Jain."],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/90472","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Kalé, Laxmikant V","Gropp, William D.","Torrellas, Josep","Panda, Dhabaleswar K."]},{"key":"dc:creator","label":"Author","values":["Jain, Nikhil"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2016-07-07T19:52:47Z","2016-02-12","2016-05"]},{"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":["Network","Communication","Parallel computing","Applications","Simulation","Modeling","Prediction","Machine learning","Topology-aware mapping"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2016, Nikhil Jain."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/90472"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Communication is a necessary but overhead inducing component of parallel programming. Its impact on application design and performance is due to several related aspects of a parallel job execution: network topology, routing protocol, suitability of algorithm being used to the network, job placement, etc. This thesis is aimed at developing an understanding of how communication plays out on networks of high performance computing systems and exploring methods that can be used to improve communication performance of large scale applications. Broadly speaking, three topics have been studied in detail in this thesis. The first of these topics is task mapping and job placement on practical installations of torus and dragonfly networks. Next, use of supervised learning algorithms for conducting diagnostic studies of how communication evolves on networks is explored. Finally, efficacy of packet-level simulations for prediction-based studies of communication performance on different networks using different network parameters is analyzed. The primary contribution of this thesis is development of scalable diagnostic and prediction methods that can assist in the process of network designing, adapting applications to future systems, and optimizing execution of applications on existing systems. These meth- ods include a supervised learning approach, a functional modeling tool (called Damselfly), and a PDES-based packet level simulator (called TraceR), all of which are described in this thesis.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2016-07-07 without embargo terms","The student, Nikhil Jain, accepted the attached license on 2016-02-09 at 17:43.","The student, Nikhil Jain, submitted this Dissertation for approval on 2016-02-09 at 17:54.","This Dissertation was approved for publication on 2016-02-12 at 15:08.","DSpace SAF Submission Ingestion Package generated from Vireo submission #9074 on 2016-07-07 at 13:26:59","Made available in DSpace on 2016-07-07T19:52:47Z (GMT). No. of bitstreams: 2 JAIN-DISSERTATION-2016.pdf: 6254901 bytes, checksum: 704bd96dbb8446a568e24a82db96e68f (MD5) LICENSE.txt: 4208 bytes, checksum: ddbbff6499598dea906076f86b514177 (MD5) Previous issue date: 2016-02-12"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Optimization of communication intensive applications on HPC networks"]}]}],"canonical_facts":{"dc:contributor":["Kalé, Laxmikant V","Gropp, William D.","Torrellas, Josep","Panda, Dhabaleswar K."],"dc:creator":["Jain, Nikhil"],"dc:date":["2016-07-07T19:52:47Z","2016-02-12","2016-05"],"dc:description":["Communication is a necessary but overhead inducing component of parallel programming. Its impact on application design and performance is due to several related aspects of a parallel job execution: network topology, routing protocol, suitability of algorithm being used to the network, job placement, etc. This thesis is aimed at developing an understanding of how communication plays out on networks of high performance computing systems and exploring methods that can be used to improve communication performance of large scale applications. Broadly speaking, three topics have been studied in detail in this thesis. The first of these topics is task mapping and job placement on practical installations of torus and dragonfly networks. Next, use of supervised learning algorithms for conducting diagnostic studies of how communication evolves on networks is explored. Finally, efficacy of packet-level simulations for prediction-based studies of communication performance on different networks using different network parameters is analyzed. The primary contribution of this thesis is development of scalable diagnostic and prediction methods that can assist in the process of network designing, adapting applications to future systems, and optimizing execution of applications on existing systems. These meth- ods include a supervised learning approach, a functional modeling tool (called Damselfly), and a PDES-based packet level simulator (called TraceR), all of which are described in this thesis.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2016-07-07 without embargo terms","The student, Nikhil Jain, accepted the attached license on 2016-02-09 at 17:43.","The student, Nikhil Jain, submitted this Dissertation for approval on 2016-02-09 at 17:54.","This Dissertation was approved for publication on 2016-02-12 at 15:08.","DSpace SAF Submission Ingestion Package generated from Vireo submission #9074 on 2016-07-07 at 13:26:59","Made available in DSpace on 2016-07-07T19:52:47Z (GMT). No. of bitstreams: 2 JAIN-DISSERTATION-2016.pdf: 6254901 bytes, checksum: 704bd96dbb8446a568e24a82db96e68f (MD5) LICENSE.txt: 4208 bytes, checksum: ddbbff6499598dea906076f86b514177 (MD5) Previous issue date: 2016-02-12"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/90472"],"dc:language":["en"],"dc:rights":["Copyright 2016, Nikhil Jain."],"dc:subject":["Network","Communication","Parallel computing","Applications","Simulation","Modeling","Prediction","Machine learning","Topology-aware mapping"],"dc:title":["Optimization of communication intensive applications on HPC networks"],"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"}