{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,14]],"date-time":"2026-03-14T09:53:44Z","timestamp":1773482024306,"version":"3.50.1"},"reference-count":36,"publisher":"Association for Computing Machinery (ACM)","issue":"11","content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["Proc. VLDB Endow."],"published-print":{"date-parts":[[2024,7]]},"abstract":"<jats:p>Dynamic Graph Neural Networks (DGNNs) have demonstrated exceptional performance at dynamic-graph analysis tasks. However, the costs exceed those incurred by other learning tasks, to the point where deployment on large-scale dynamic graphs is infeasible. Existing distributed frameworks that facilitate DGNN training are in their early stages and experience challenges such as communication bottlenecks, imbalanced workloads, and GPU memory overflow.<\/jats:p>\n          <jats:p>We introduce DynaHB, a distributed framework for DGNN training using so-called Hybrid Batches. DynaHB reduces communication by means of vertex caching, and it ensures even data and workload distribution by means of load-aware vertex partitioning. DyanHB also features a novel hybrid-batch training mode that combines vertex-batch and snapshot-batch techniques, thereby reducing training time and GPU memory usage. Next, to further enhance the hybrid batch based approach, DynaHB integrates a reinforcement learning-based batch adjuster and a pipelined batch generator with a batch reservoir to reduce the cost of generating hybrid batches. Extensive experiments show that DynaHB is capable of up to a 93\u00d7 and an average of 8.06\u00d7 speedups over the state-of-the-art training framework.<\/jats:p>","DOI":"10.14778\/3681954.3682008","type":"journal-article","created":{"date-parts":[[2024,8,30]],"date-time":"2024-08-30T16:23:36Z","timestamp":1725035016000},"page":"3388-3401","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":5,"title":["DynaHB: A Communication-Avoiding Asynchronous Distributed Framework with Hybrid Batches for Dynamic GNN Training"],"prefix":"10.14778","volume":"17","author":[{"given":"Zhen","family":"Song","sequence":"first","affiliation":[{"name":"Northeastern Univ., China"}]},{"given":"Yu","family":"Gu","sequence":"additional","affiliation":[{"name":"Northeastern Univ., China"}]},{"given":"Qing","family":"Sun","sequence":"additional","affiliation":[{"name":"Northeastern Univ., China"}]},{"given":"Tianyi","family":"Li","sequence":"additional","affiliation":[{"name":"Aalborg Univ., Denmark"}]},{"given":"Yanfeng","family":"Zhang","sequence":"additional","affiliation":[{"name":"Northeastern Univ., China"}]},{"given":"Yushuai","family":"Li","sequence":"additional","affiliation":[{"name":"Aalborg Univ., Denmark"}]},{"given":"Christian S.","family":"Jensen","sequence":"additional","affiliation":[{"name":"Aalborg Univ., Denmark"}]},{"given":"Ge","family":"Yu","sequence":"additional","affiliation":[{"name":"Northeastern Univ., China"}]}],"member":"320","published-online":{"date-parts":[[2024,8,30]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1137\/0608024"},{"key":"e_1_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-39890-5_6"},{"key":"e_1_2_1_3_1","unstructured":"Hans L Bodlaender et al. 1992. A tourist guide through treewidth. (1992)."},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00778-012-0274-x"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1145\/3448016.3459245"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1137\/S0097539702403098"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-15364-8_36"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-12038-1_20"},{"key":"e_1_2_1_9_1","doi-asserted-by":"crossref","unstructured":"Edsger W Dijkstra. 2022. A note on two problems in connexion with graphs. In Edsger Wybe Dijkstra: His Life Work and Legacy. 287--290.","DOI":"10.1145\/3544585.3544600"},{"key":"e_1_2_1_10_1","unstructured":"Alexandros Efentakis. 2016. Scalable Public Transportation Queries on the Database.. In EDBT. 527--538."},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE53745.2022.00043"},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jalgor.2004.05.002"},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE.2019.00118"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/SKG49510.2019.00020"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1109\/SRDSW.2014.17"},{"key":"e_1_2_1_16_1","volume-title":"ROAD: A new spatial object search framework for road networks","author":"Lee Ken CK","year":"2010","unstructured":"Ken CK Lee, Wang-Chien Lee, Baihua Zheng, and Yuan Tian. 2010. ROAD: A new spatial object search framework for road networks. IEEE transactions on knowledge and data engineering 24, 3 (2010), 547--560."},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00778-022-00758-w"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/TrustCom53373.2021.00151"},{"key":"e_1_2_1_19_1","first-page":"300","article-title":"Fastest path query answering using time-dependent hop-labeling in road network","volume":"34","author":"Li Lei","year":"2020","unstructured":"Lei Li, Sibo Wang, and Xiaofang Zhou. 2020. Fastest path query answering using time-dependent hop-labeling in road network. IEEE Transactions on Knowledge and Data Engineering 34, 1 (2020), 300--313.","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3389748"},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/3187009.3177736"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/3183713.3196913"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3389746"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/B978-012722442-8\/50076-8"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/0095-8956(84)90013-3"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/1376616.1376623"},{"key":"e_1_2_1_27_1","volume-title":"Equitable Public Bus Network Optimization for Social Good: A Case Study of Singapore. In 2022 ACM Conference on Fairness, Accountability, and Transparency. 278--288","author":"Tedjopurnomo David","year":"2022","unstructured":"David Tedjopurnomo, Zhifeng Bao, Farhana Choudhury, Hui Luo, and AK Qin. 2022. Equitable Public Bus Network Optimization for Social Good: A Case Study of Singapore. In 2022 ACM Conference on Fairness, Accountability, and Transparency. 278--288."},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/2723372.2749456"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1145\/3448016.3457247"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1145\/1807167.1807181"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcss.2015.06.008"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1109\/CSB.2005.9"},{"key":"e_1_2_1_33_1","doi-asserted-by":"crossref","unstructured":"Yajun Yang Hanxiao Li Junhu Wang Qinghua Hu Xin Wang Muxi Leng et al. 2019. A novel index method for k nearest object query over time-dependent road networks. Complexity 2019 (2019).","DOI":"10.1155\/2019\/4829164"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.14778\/3551793.3551820"},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00778-023-00825-w"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2015.2399306"}],"container-title":["Proceedings of the VLDB Endowment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.14778\/3681954.3682008","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,4]],"date-time":"2024-09-04T18:35:29Z","timestamp":1725474929000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.14778\/3681954.3682008"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,7]]},"references-count":36,"journal-issue":{"issue":"11","published-print":{"date-parts":[[2024,7]]}},"alternative-id":["10.14778\/3681954.3682008"],"URL":"https:\/\/doi.org\/10.14778\/3681954.3682008","relation":{},"ISSN":["2150-8097"],"issn-type":[{"value":"2150-8097","type":"print"}],"subject":[],"published":{"date-parts":[[2024,7]]},"assertion":[{"value":"2024-08-30","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}