{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/pgt-i-scaling-spatiotemporal-gnns-with-memory","title":"PGT-I: Scaling Spatiotemporal GNNs with Memory-Efficient Distributed Training","arxiv_id":"2507.11683","date":"2025-07-15","proceeding":null,"authors":["Seth Ockerman","Amal Gueroudji","Tanwi Mallick","Yixuan He","Line Pouchard","Robert Ross","Shivaram Venkataraman"],"abstract":"Spatiotemporal graph neural networks (ST-GNNs) are powerful tools for modeling spatial and temporal data dependencies. However, their applications have been limited primarily to small-scale datasets because of memory constraints. While distributed training offers a solution, current frameworks lack support for spatiotemporal models and overlook the properties of spatiotemporal data. Informed by a scaling study on a large-scale workload, we present PyTorch Geometric Temporal Index (PGT-I), an extension to PyTorch Geometric Temporal that integrates distributed data parallel training and two novel strategies: index-batching and distributed-index-batching. Our index techniques exploit spatiotemporal structure to construct snapshots dynamically at runtime, significantly reducing memory overhead, while distributed-index-batching extends this approach by enabling scalable processing across multiple GPUs. Our techniques enable the first-ever training of an ST-GNN on the entire PeMS dataset without graph partitioning, reducing peak memory usage by up to 89\\% and achieving up to a 13.1x speedup over standard DDP with 128 GPUs.","url_abs":"https://arxiv.org/abs/2507.11683v1","url_pdf":"https://arxiv.org/pdf/2507.11683v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"pgt-i-scaling-spatiotemporal-gnns-with-memory","repo_url":"https://github.com/benedekrozemberczki/pytorch_geometric_temporal","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"pgt-i-scaling-spatiotemporal-gnns-with-memory","repo_url":"https://github.com/uw-mad-dash/pgt_index","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"graph-partitioning","task_name":"graph partitioning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}