{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:BR75OIUY7WDYHPO53K7WUNQ4NZ","short_pith_number":"pith:BR75OIUY","canonical_record":{"source":{"id":"1907.13257","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2019-07-30T23:20:50Z","cross_cats_sorted":["cs.AI","cs.DC","stat.ML"],"title_canon_sha256":"43ac329a7fd38ac35d9176973eebda4fd46590aeadbc58fff8e8c1e680c989f4","abstract_canon_sha256":"d1161812bdf97d61dc95e12e809f35203eebd989122dc563ccc21bce1aaab232"},"schema_version":"1.0"},"canonical_sha256":"0c7fd72298fd8783bddddabf6a361c6e7ae24b3654f11ff6a67888be65f2523c","source":{"kind":"arxiv","id":"1907.13257","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.13257","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"arxiv_version","alias_value":"1907.13257v1","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.13257","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"pith_short_12","alias_value":"BR75OIUY7WDY","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"pith_short_16","alias_value":"BR75OIUY7WDYHPO5","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"pith_short_8","alias_value":"BR75OIUY","created_at":"2026-07-05T05:13:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:BR75OIUY7WDYHPO53K7WUNQ4NZ","target":"record","payload":{"canonical_record":{"source":{"id":"1907.13257","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2019-07-30T23:20:50Z","cross_cats_sorted":["cs.AI","cs.DC","stat.ML"],"title_canon_sha256":"43ac329a7fd38ac35d9176973eebda4fd46590aeadbc58fff8e8c1e680c989f4","abstract_canon_sha256":"d1161812bdf97d61dc95e12e809f35203eebd989122dc563ccc21bce1aaab232"},"schema_version":"1.0"},"canonical_sha256":"0c7fd72298fd8783bddddabf6a361c6e7ae24b3654f11ff6a67888be65f2523c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:13:27.052835Z","signature_b64":"xfHWL3OkolFKPwJYGc6123BEFCjtWZFD9qILaldbJLs3LJRv0YzHsjS3ul9ClL1xJc/Ni421fvgOPWASgQejDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c7fd72298fd8783bddddabf6a361c6e7ae24b3654f11ff6a67888be65f2523c","last_reissued_at":"2026-07-05T05:13:27.052423Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:13:27.052423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1907.13257","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:13:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QKimZnSYQf+2p8Bi+sle08iBm8265fiSg7RZiSG/IGPajZp+r6RLRO15OHqaIsQlvJZF7s83Akoyqj9Osre6AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T17:11:15.377223Z"},"content_sha256":"6fdcaaebde1a456ee3938bf1778aa2a48bf950c316fc411c8b4bcb6a15585758","schema_version":"1.0","event_id":"sha256:6fdcaaebde1a456ee3938bf1778aa2a48bf950c316fc411c8b4bcb6a15585758"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:BR75OIUY7WDYHPO53K7WUNQ4NZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimizing Multi-GPU Parallelization Strategies for Deep Learning Training","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.DC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Arslan Zulfiqar, David Nellans, Eiman Ebrahimi, Puneet Gupta, Saptadeep Pal, Szymon Migacz, Victor Zhang, Yaosheng Fu","submitted_at":"2019-07-30T23:20:50Z","abstract_excerpt":"Deploying deep learning (DL) models across multiple compute devices to train large and complex models continues to grow in importance because of the demand for faster and more frequent training. Data parallelism (DP) is the most widely used parallelization strategy, but as the number of devices in data parallel training grows, so does the communication overhead between devices. Additionally, a larger aggregate batch size per step leads to statistical efficiency loss, i.e., a larger number of epochs are required to converge to a desired accuracy. These factors affect overall training time and b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.13257","kind":"arxiv","version":1},"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/1907.13257/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:13:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"e7ByZAGHxdkS8RcVW8wvtVtGHNJslyaCalUXkwkdJ6TO7w+1WLaTJa6Y6YHSzWGXZS7wga3rPrC9RSu1QU+BAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T17:11:15.377795Z"},"content_sha256":"22bc30fa1f968e80e55beb5e1ecb788a49c214c9b497174c45ea04d49fffff0f","schema_version":"1.0","event_id":"sha256:22bc30fa1f968e80e55beb5e1ecb788a49c214c9b497174c45ea04d49fffff0f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BR75OIUY7WDYHPO53K7WUNQ4NZ/bundle.json","state_url":"https://pith.science/pith/BR75OIUY7WDYHPO53K7WUNQ4NZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BR75OIUY7WDYHPO53K7WUNQ4NZ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-21T17:11:15Z","links":{"resolver":"https://pith.science/pith/BR75OIUY7WDYHPO53K7WUNQ4NZ","bundle":"https://pith.science/pith/BR75OIUY7WDYHPO53K7WUNQ4NZ/bundle.json","state":"https://pith.science/pith/BR75OIUY7WDYHPO53K7WUNQ4NZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BR75OIUY7WDYHPO53K7WUNQ4NZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:BR75OIUY7WDYHPO53K7WUNQ4NZ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d1161812bdf97d61dc95e12e809f35203eebd989122dc563ccc21bce1aaab232","cross_cats_sorted":["cs.AI","cs.DC","stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2019-07-30T23:20:50Z","title_canon_sha256":"43ac329a7fd38ac35d9176973eebda4fd46590aeadbc58fff8e8c1e680c989f4"},"schema_version":"1.0","source":{"id":"1907.13257","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.13257","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"arxiv_version","alias_value":"1907.13257v1","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.13257","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"pith_short_12","alias_value":"BR75OIUY7WDY","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"pith_short_16","alias_value":"BR75OIUY7WDYHPO5","created_at":"2026-07-05T05:13:27Z"},{"alias_kind":"pith_short_8","alias_value":"BR75OIUY","created_at":"2026-07-05T05:13:27Z"}],"graph_snapshots":[{"event_id":"sha256:22bc30fa1f968e80e55beb5e1ecb788a49c214c9b497174c45ea04d49fffff0f","target":"graph","created_at":"2026-07-05T05:13:27Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/1907.13257/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deploying deep learning (DL) models across multiple compute devices to train large and complex models continues to grow in importance because of the demand for faster and more frequent training. Data parallelism (DP) is the most widely used parallelization strategy, but as the number of devices in data parallel training grows, so does the communication overhead between devices. Additionally, a larger aggregate batch size per step leads to statistical efficiency loss, i.e., a larger number of epochs are required to converge to a desired accuracy. These factors affect overall training time and b","authors_text":"Arslan Zulfiqar, David Nellans, Eiman Ebrahimi, Puneet Gupta, Saptadeep Pal, Szymon Migacz, Victor Zhang, Yaosheng Fu","cross_cats":["cs.AI","cs.DC","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2019-07-30T23:20:50Z","title":"Optimizing Multi-GPU Parallelization Strategies for Deep Learning Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.13257","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6fdcaaebde1a456ee3938bf1778aa2a48bf950c316fc411c8b4bcb6a15585758","target":"record","created_at":"2026-07-05T05:13:27Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"d1161812bdf97d61dc95e12e809f35203eebd989122dc563ccc21bce1aaab232","cross_cats_sorted":["cs.AI","cs.DC","stat.ML"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2019-07-30T23:20:50Z","title_canon_sha256":"43ac329a7fd38ac35d9176973eebda4fd46590aeadbc58fff8e8c1e680c989f4"},"schema_version":"1.0","source":{"id":"1907.13257","kind":"arxiv","version":1}},"canonical_sha256":"0c7fd72298fd8783bddddabf6a361c6e7ae24b3654f11ff6a67888be65f2523c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0c7fd72298fd8783bddddabf6a361c6e7ae24b3654f11ff6a67888be65f2523c","first_computed_at":"2026-07-05T05:13:27.052423Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:13:27.052423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xfHWL3OkolFKPwJYGc6123BEFCjtWZFD9qILaldbJLs3LJRv0YzHsjS3ul9ClL1xJc/Ni421fvgOPWASgQejDA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:13:27.052835Z","signed_message":"canonical_sha256_bytes"},"source_id":"1907.13257","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6fdcaaebde1a456ee3938bf1778aa2a48bf950c316fc411c8b4bcb6a15585758","sha256:22bc30fa1f968e80e55beb5e1ecb788a49c214c9b497174c45ea04d49fffff0f"],"state_sha256":"1534baf2947e3d0e9bd0e66e0bb3c42b16234a41abbe8710983a09fc48a535f9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WC05CD7b9GHpr4VRm1oGWcSwNRLJWQyzyMrkKS33/Cp9Ms5Fx2bj8kEjGgRfkVnAkB2y7zu4QzDKImnIIziABg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T17:11:15.382889Z","bundle_sha256":"f9f2252a11cd6561cbcd73539a1565e726170cf8665d924600243c15e27e9bcd"}}