{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ODEBBXEWXLC6G65Y6GLHGCG2F4","short_pith_number":"pith:ODEBBXEW","schema_version":"1.0","canonical_sha256":"70c810dc96bac5e37bb8f1967308da2f02556e2e1f9566c8b349af87f3978d55","source":{"kind":"arxiv","id":"2112.01075","version":2},"attestation_state":"computed","paper":{"title":"Memory-efficient array redistribution through portable collective communication","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.PL"],"primary_cat":"cs.DC","authors_text":"Adam Paszke, Dimitrios Vytiniotis, Georg Stefan Schmid, Norman A. Rink","submitted_at":"2021-12-02T09:32:07Z","abstract_excerpt":"Modern large-scale deep learning workloads highlight the need for parallel execution across many devices in order to fit model data into hardware accelerator memories. In these settings, array redistribution may be required during a computation, but can also become a bottleneck if not done efficiently. In this paper we address the problem of redistributing multi-dimensional array data in SPMD computations, the most prevalent form of parallelism in deep learning. We present a type-directed approach to synthesizing array redistributions as sequences of MPI-style collective operations. We prove f"},"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":"2112.01075","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2021-12-02T09:32:07Z","cross_cats_sorted":["cs.LG","cs.PL"],"title_canon_sha256":"04a647a636ea155cd0f5622a78ee2a554c92f3584f5e3f58ea30bb2f7269b675","abstract_canon_sha256":"3c5544e770bb0fa40d9d9fef2f8e65bf39313738c9ba23f15a4ad06e31859d4b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:19:51.553588Z","signature_b64":"EnsTkZKXBti45uQHqLsy9e5c26ufM/RUqaWY52gDUXYeJcNCRoIna4k6iygrdXkvdlXiRngUlsbeIBe1CZ8rDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"70c810dc96bac5e37bb8f1967308da2f02556e2e1f9566c8b349af87f3978d55","last_reissued_at":"2026-07-05T05:19:51.553169Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:19:51.553169Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Memory-efficient array redistribution through portable collective communication","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.PL"],"primary_cat":"cs.DC","authors_text":"Adam Paszke, Dimitrios Vytiniotis, Georg Stefan Schmid, Norman A. Rink","submitted_at":"2021-12-02T09:32:07Z","abstract_excerpt":"Modern large-scale deep learning workloads highlight the need for parallel execution across many devices in order to fit model data into hardware accelerator memories. In these settings, array redistribution may be required during a computation, but can also become a bottleneck if not done efficiently. In this paper we address the problem of redistributing multi-dimensional array data in SPMD computations, the most prevalent form of parallelism in deep learning. We present a type-directed approach to synthesizing array redistributions as sequences of MPI-style collective operations. We prove f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.01075","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/2112.01075/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":"2112.01075","created_at":"2026-07-05T05:19:51.553225+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.01075v2","created_at":"2026-07-05T05:19:51.553225+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.01075","created_at":"2026-07-05T05:19:51.553225+00:00"},{"alias_kind":"pith_short_12","alias_value":"ODEBBXEWXLC6","created_at":"2026-07-05T05:19:51.553225+00:00"},{"alias_kind":"pith_short_16","alias_value":"ODEBBXEWXLC6G65Y","created_at":"2026-07-05T05:19:51.553225+00:00"},{"alias_kind":"pith_short_8","alias_value":"ODEBBXEW","created_at":"2026-07-05T05:19:51.553225+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/ODEBBXEWXLC6G65Y6GLHGCG2F4","json":"https://pith.science/pith/ODEBBXEWXLC6G65Y6GLHGCG2F4.json","graph_json":"https://pith.science/api/pith-number/ODEBBXEWXLC6G65Y6GLHGCG2F4/graph.json","events_json":"https://pith.science/api/pith-number/ODEBBXEWXLC6G65Y6GLHGCG2F4/events.json","paper":"https://pith.science/paper/ODEBBXEW"},"agent_actions":{"view_html":"https://pith.science/pith/ODEBBXEWXLC6G65Y6GLHGCG2F4","download_json":"https://pith.science/pith/ODEBBXEWXLC6G65Y6GLHGCG2F4.json","view_paper":"https://pith.science/paper/ODEBBXEW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.01075&json=true","fetch_graph":"https://pith.science/api/pith-number/ODEBBXEWXLC6G65Y6GLHGCG2F4/graph.json","fetch_events":"https://pith.science/api/pith-number/ODEBBXEWXLC6G65Y6GLHGCG2F4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ODEBBXEWXLC6G65Y6GLHGCG2F4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ODEBBXEWXLC6G65Y6GLHGCG2F4/action/storage_attestation","attest_author":"https://pith.science/pith/ODEBBXEWXLC6G65Y6GLHGCG2F4/action/author_attestation","sign_citation":"https://pith.science/pith/ODEBBXEWXLC6G65Y6GLHGCG2F4/action/citation_signature","submit_replication":"https://pith.science/pith/ODEBBXEWXLC6G65Y6GLHGCG2F4/action/replication_record"}},"created_at":"2026-07-05T05:19:51.553225+00:00","updated_at":"2026-07-05T05:19:51.553225+00:00"}