{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/127395"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/127395","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Designing scalable large-scale storage-based GNN framework by efficiently leveraging heterogeneous hardware resources","abstract":"The student, Jeongmin Park, submitted this Dissertation for approval on 2024-12-03 at 22:05.","abstract_html":"The student, Jeongmin Park, submitted this Dissertation for approval on 2024-12-03 at 22:05.","abstract_has_math":false,"creators":["Park, Jeongmin"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Electrical & Computer Engr","degree_department":null,"school":null,"contributors":["Hwu, Wen-mei","Patel, Sanjay","Chen, Deming","Lumetta, Steven S"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-12-04","date_published":"2024-12-04","updated_at":"2026-07-22T22:25:04Z","subjects":["Storage","Large-scale Ml","Distributed System","Gpu","Parallel Computing","Gnn"],"languages":["eng","en"],"rights":["Copyright 2024 Jeongmin Park"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2142/127395","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Hwu, Wen-mei","Patel, Sanjay","Chen, Deming","Lumetta, Steven S"]},{"key":"dc:creator","label":"Author","values":["Park, Jeongmin"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2024-12-04","2024-12"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical & Computer Engr"]},{"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":["Storage","Large-scale Ml","Distributed System","Gpu","Parallel Computing","Gnn"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng","en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2024 Jeongmin Park"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/2142/127395"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["The student, Jeongmin Park, submitted this Dissertation for approval on 2024-12-03 at 22:05.","This Dissertation was approved for publication on 2024-12-04 at 14:55.","DSpace SAF Submission Ingestion Package generated from Vireo submission #21489 on 2025-03-28 at 14:44:39","Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2026-12-01","The student, Jeongmin Park, accepted the attached license on 2024-12-03 at 21:49.","Graph Neural Networks (GNNs) are widely used today in applications such as recommendation systems, fraud detection, and node/link classification tasks. As real-world graphs and embeddings used for GNN training continue to grow in scale, their memory footprint often exceeds the memory capacities of GPUs, creating a bottleneck in efficient training. Traditional GNN training frameworks address limited memory by either storing feature data in external storage and fetching it on-demand or sharding the graph across multiple GPUs and transferring data as needed. However, the first approach suffers from high storage latency, while the second is burdened by the high computational costs of graph partitioning, excessive inter-GPU communication, and increased total cost of ownership. To address these challenges, this dissertation introduces three storage-based GNN frameworks—GIDS, LSM-GNN, and SSD-GNN—that span single-GPU, multi-GPU, and multi-node environments, respectively. GIDS accelerates single-GPU GNN training by leveraging GPU thread parallelism to hide storage latency. LSM-GNN extends this to a multi-GPU setting by implementing a system-wide shared cache using NVLink, optimizing memory bandwidth and cache hit rates without graph partitioning. SSD-GNN further scales GNN training across multiple nodes by integrating GPU-initiated direct storage access and a distributed caching protocol to reduce data movement overhead across nodes. These frameworks effectively utilize heterogeneous hardware resources such as SSDs, CPU memory, GPU parallelism, and GPU PCIe bandwidth to optimize data transfer operations and hide storage latency. Prototypes of these systems demonstrate significant improvements in GNN training performance and reduced total cost of ownership compared to state-of-the-art systems, marking a new pathway for scalable, storage-efficient GNN training across a range of compute environments."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Designing scalable large-scale storage-based GNN framework by efficiently leveraging heterogeneous hardware resources"]}]}],"canonical_facts":{"dc:contributor":["Hwu, Wen-mei","Patel, Sanjay","Chen, Deming","Lumetta, Steven S"],"dc:creator":["Park, Jeongmin"],"dc:date":["2024-12-04","2024-12"],"dc:description":["The student, Jeongmin Park, submitted this Dissertation for approval on 2024-12-03 at 22:05.","This Dissertation was approved for publication on 2024-12-04 at 14:55.","DSpace SAF Submission Ingestion Package generated from Vireo submission #21489 on 2025-03-28 at 14:44:39","Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2026-12-01","The student, Jeongmin Park, accepted the attached license on 2024-12-03 at 21:49.","Graph Neural Networks (GNNs) are widely used today in applications such as recommendation systems, fraud detection, and node/link classification tasks. As real-world graphs and embeddings used for GNN training continue to grow in scale, their memory footprint often exceeds the memory capacities of GPUs, creating a bottleneck in efficient training. Traditional GNN training frameworks address limited memory by either storing feature data in external storage and fetching it on-demand or sharding the graph across multiple GPUs and transferring data as needed. However, the first approach suffers from high storage latency, while the second is burdened by the high computational costs of graph partitioning, excessive inter-GPU communication, and increased total cost of ownership. To address these challenges, this dissertation introduces three storage-based GNN frameworks—GIDS, LSM-GNN, and SSD-GNN—that span single-GPU, multi-GPU, and multi-node environments, respectively. GIDS accelerates single-GPU GNN training by leveraging GPU thread parallelism to hide storage latency. LSM-GNN extends this to a multi-GPU setting by implementing a system-wide shared cache using NVLink, optimizing memory bandwidth and cache hit rates without graph partitioning. SSD-GNN further scales GNN training across multiple nodes by integrating GPU-initiated direct storage access and a distributed caching protocol to reduce data movement overhead across nodes. These frameworks effectively utilize heterogeneous hardware resources such as SSDs, CPU memory, GPU parallelism, and GPU PCIe bandwidth to optimize data transfer operations and hide storage latency. Prototypes of these systems demonstrate significant improvements in GNN training performance and reduced total cost of ownership compared to state-of-the-art systems, marking a new pathway for scalable, storage-efficient GNN training across a range of compute environments."],"dc:format":["application/pdf"],"dc:identifier":["https://hdl.handle.net/2142/127395"],"dc:language":["eng","en"],"dc:rights":["Copyright 2024 Jeongmin Park"],"dc:subject":["Storage","Large-scale Ml","Distributed System","Gpu","Parallel Computing","Gnn"],"dc:title":["Designing scalable large-scale storage-based GNN framework by efficiently leveraging heterogeneous hardware resources"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Electrical & Computer Engr"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:25:04Z"}