{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/121343"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/121343","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Auto-parallelization of machine-learning dataflow graphs for CPU multicores","abstract":"Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2025-08-01","abstract_html":"Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2025-08-01","abstract_has_math":false,"creators":["Das, Srinjoy"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Rauchwerger, Lawrence"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2023,"date_issued":"2023-08","date_published":"2023-08","updated_at":"2026-07-22T22:24:57Z","subjects":["Parallelization","Clustering","Machine Learning","Graph Optimization","Compiler Optimization","Dataflow Graph","Inference","Multicores","Pytorch"],"languages":["en","eng"],"rights":["Copyright 2023 Srinjoy Das"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2142/121343","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Rauchwerger, Lawrence"]},{"key":"dc:creator","label":"Author","values":["Das, Srinjoy"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2023-08","2023-07-17"]},{"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":["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":["Parallelization","Clustering","Machine Learning","Graph Optimization","Compiler Optimization","Dataflow Graph","Inference","Multicores","Pytorch"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en","eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2023 Srinjoy Das"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/2142/121343"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2025-08-01","The student, Srinjoy Das, accepted the attached license on 2023-07-08 at 22:01.","The student, Srinjoy Das, submitted this Thesis for approval on 2023-07-08 at 22:06.","This Thesis was approved for publication on 2023-07-17 at 09:08.","DSpace SAF Submission Ingestion Package generated from Vireo submission #19566 on 2023-12-04 at 17:18:24","Several methods exist today to accelerate Machine Learning(ML)/Deep-Learning(DL) model performance for training and inference. However, modern techniques that rely on various graph and operator parallelism methodologies rely on search space optimizations which are costly in terms of power and hardware usage. Especially in the case of inference, when the batch size is 1 and execution is on Central Processing Units (CPUs) or at the edge, current techniques can become costly, complicated or inapplicable. To ameliorate this, we present a Critical-Path-based Linear Clustering approach to exploit inherent parallel paths in ML dataflow graphs. We augment this with a new hyperclustering mechanism for small batch sizes > 1 which may be typical in inference scenarios. Our task parallelization approach further optimizes the structure of graphs via cloning and simplifies them via dead-code elimination. Contrary to other work, we generate readable and executable parallel Pytorch+Python code from input ONNX models via a new tool that we have built called Ramiel which allows us to benefit from other downstream acceleration techniques like intra-op parallelism and potentially pipeline parallelism. Our preliminary results on several ML graphs demonstrate up to 1.9× speedup over serial execution and outperform some of the current mechanisms in both compile and runtimes. Lastly, our methods are lightweight and fast enough so that they can be used effectively for Artificial Intelligence (AI) at the edge."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Auto-parallelization of machine-learning dataflow graphs for CPU multicores"]}]}],"canonical_facts":{"dc:contributor":["Rauchwerger, Lawrence"],"dc:creator":["Das, Srinjoy"],"dc:date":["2023-08","2023-07-17"],"dc:description":["Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2025-08-01","The student, Srinjoy Das, accepted the attached license on 2023-07-08 at 22:01.","The student, Srinjoy Das, submitted this Thesis for approval on 2023-07-08 at 22:06.","This Thesis was approved for publication on 2023-07-17 at 09:08.","DSpace SAF Submission Ingestion Package generated from Vireo submission #19566 on 2023-12-04 at 17:18:24","Several methods exist today to accelerate Machine Learning(ML)/Deep-Learning(DL) model performance for training and inference. However, modern techniques that rely on various graph and operator parallelism methodologies rely on search space optimizations which are costly in terms of power and hardware usage. Especially in the case of inference, when the batch size is 1 and execution is on Central Processing Units (CPUs) or at the edge, current techniques can become costly, complicated or inapplicable. To ameliorate this, we present a Critical-Path-based Linear Clustering approach to exploit inherent parallel paths in ML dataflow graphs. We augment this with a new hyperclustering mechanism for small batch sizes > 1 which may be typical in inference scenarios. Our task parallelization approach further optimizes the structure of graphs via cloning and simplifies them via dead-code elimination. Contrary to other work, we generate readable and executable parallel Pytorch+Python code from input ONNX models via a new tool that we have built called Ramiel which allows us to benefit from other downstream acceleration techniques like intra-op parallelism and potentially pipeline parallelism. Our preliminary results on several ML graphs demonstrate up to 1.9× speedup over serial execution and outperform some of the current mechanisms in both compile and runtimes. Lastly, our methods are lightweight and fast enough so that they can be used effectively for Artificial Intelligence (AI) at the edge."],"dc:format":["application/pdf"],"dc:identifier":["https://hdl.handle.net/2142/121343"],"dc:language":["en","eng"],"dc:rights":["Copyright 2023 Srinjoy Das"],"dc:subject":["Parallelization","Clustering","Machine Learning","Graph Optimization","Compiler Optimization","Dataflow Graph","Inference","Multicores","Pytorch"],"dc:title":["Auto-parallelization of machine-learning dataflow graphs for CPU multicores"],"dc:type":["text"],"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":"2026-07-22T22:24:57Z"}