{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:LLASAULE7PRCQ4OVCY7C2OL5YS","short_pith_number":"pith:LLASAULE","schema_version":"1.0","canonical_sha256":"5ac1205164fbe22871d5163e2d397dc493e1018dbe3e28de1b12c0e53b18b3ce","source":{"kind":"arxiv","id":"2105.11118","version":2},"attestation_state":"computed","paper":{"title":"Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Guanzhou Hu, Guoqing Harry Xu, Jinliang Wei, John Thorpe, Jonathan Eyolfson, Keval Vora, Miryung Kim, Ravi Netravali, Shen Teng, Yifan Qiao, Zhihao Jia","submitted_at":"2021-05-24T06:49:08Z","abstract_excerpt":"A graph neural network (GNN) enables deep learning on structured graph data. There are two major GNN training obstacles: 1) it relies on high-end servers with many GPUs which are expensive to purchase and maintain, and 2) limited memory on GPUs cannot scale to today's billion-edge graphs. This paper presents Dorylus: a distributed system for training GNNs. Uniquely, Dorylus can take advantage of serverless computing to increase scalability at a low cost.\n  The key insight guiding our design is computation separation. Computation separation makes it possible to construct a deep, bounded-asynchr"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2105.11118","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2021-05-24T06:49:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e72e370cfaaa6fda455d412a668a28dcfef2a9112e693cfc947ab39d6b840a7a","abstract_canon_sha256":"bfa929ed977e7abb9009bb88bb697a1033e5398bd80aad320f70063cd4969091"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:42:52.483189Z","signature_b64":"MlEcbFMFNRRejy/DKyqV2Jnw0m5kAZ16Y6F1+r/pAw29irfw5FHpeLOR0CFrgOGXaqVwzGrKWPiz4Dkhaj95AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5ac1205164fbe22871d5163e2d397dc493e1018dbe3e28de1b12c0e53b18b3ce","last_reissued_at":"2026-07-05T02:42:52.482692Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:42:52.482692Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Guanzhou Hu, Guoqing Harry Xu, Jinliang Wei, John Thorpe, Jonathan Eyolfson, Keval Vora, Miryung Kim, Ravi Netravali, Shen Teng, Yifan Qiao, Zhihao Jia","submitted_at":"2021-05-24T06:49:08Z","abstract_excerpt":"A graph neural network (GNN) enables deep learning on structured graph data. There are two major GNN training obstacles: 1) it relies on high-end servers with many GPUs which are expensive to purchase and maintain, and 2) limited memory on GPUs cannot scale to today's billion-edge graphs. This paper presents Dorylus: a distributed system for training GNNs. Uniquely, Dorylus can take advantage of serverless computing to increase scalability at a low cost.\n  The key insight guiding our design is computation separation. Computation separation makes it possible to construct a deep, bounded-asynchr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.11118","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2105.11118/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2105.11118","created_at":"2026-07-05T02:42:52.482764+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.11118v2","created_at":"2026-07-05T02:42:52.482764+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.11118","created_at":"2026-07-05T02:42:52.482764+00:00"},{"alias_kind":"pith_short_12","alias_value":"LLASAULE7PRC","created_at":"2026-07-05T02:42:52.482764+00:00"},{"alias_kind":"pith_short_16","alias_value":"LLASAULE7PRCQ4OV","created_at":"2026-07-05T02:42:52.482764+00:00"},{"alias_kind":"pith_short_8","alias_value":"LLASAULE","created_at":"2026-07-05T02:42:52.482764+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LLASAULE7PRCQ4OVCY7C2OL5YS","json":"https://pith.science/pith/LLASAULE7PRCQ4OVCY7C2OL5YS.json","graph_json":"https://pith.science/api/pith-number/LLASAULE7PRCQ4OVCY7C2OL5YS/graph.json","events_json":"https://pith.science/api/pith-number/LLASAULE7PRCQ4OVCY7C2OL5YS/events.json","paper":"https://pith.science/paper/LLASAULE"},"agent_actions":{"view_html":"https://pith.science/pith/LLASAULE7PRCQ4OVCY7C2OL5YS","download_json":"https://pith.science/pith/LLASAULE7PRCQ4OVCY7C2OL5YS.json","view_paper":"https://pith.science/paper/LLASAULE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.11118&json=true","fetch_graph":"https://pith.science/api/pith-number/LLASAULE7PRCQ4OVCY7C2OL5YS/graph.json","fetch_events":"https://pith.science/api/pith-number/LLASAULE7PRCQ4OVCY7C2OL5YS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LLASAULE7PRCQ4OVCY7C2OL5YS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LLASAULE7PRCQ4OVCY7C2OL5YS/action/storage_attestation","attest_author":"https://pith.science/pith/LLASAULE7PRCQ4OVCY7C2OL5YS/action/author_attestation","sign_citation":"https://pith.science/pith/LLASAULE7PRCQ4OVCY7C2OL5YS/action/citation_signature","submit_replication":"https://pith.science/pith/LLASAULE7PRCQ4OVCY7C2OL5YS/action/replication_record"}},"created_at":"2026-07-05T02:42:52.482764+00:00","updated_at":"2026-07-05T02:42:52.482764+00:00"}