{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:3YN673EVFOVJAYQNLYBADBBI3I","short_pith_number":"pith:3YN673EV","canonical_record":{"source":{"id":"2201.07821","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-01-19T19:05:42Z","cross_cats_sorted":["cs.PF"],"title_canon_sha256":"f5bc5ef2532bd9243ab9d2d79e69b40351de8d876652fe2f17c0db919cfcf1cc","abstract_canon_sha256":"64b5d9c68555d0ccf16c8694a29931dbb4aacff15629b3c76a9b12c768d4d846"},"schema_version":"1.0"},"canonical_sha256":"de1befec952baa90620d5e02018428da3ef44ee4d913af7ec44077b7226325ef","source":{"kind":"arxiv","id":"2201.07821","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.07821","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"arxiv_version","alias_value":"2201.07821v2","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.07821","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"pith_short_12","alias_value":"3YN673EVFOVJ","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"pith_short_16","alias_value":"3YN673EVFOVJAYQN","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"pith_short_8","alias_value":"3YN673EV","created_at":"2026-07-05T05:16:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:3YN673EVFOVJAYQNLYBADBBI3I","target":"record","payload":{"canonical_record":{"source":{"id":"2201.07821","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-01-19T19:05:42Z","cross_cats_sorted":["cs.PF"],"title_canon_sha256":"f5bc5ef2532bd9243ab9d2d79e69b40351de8d876652fe2f17c0db919cfcf1cc","abstract_canon_sha256":"64b5d9c68555d0ccf16c8694a29931dbb4aacff15629b3c76a9b12c768d4d846"},"schema_version":"1.0"},"canonical_sha256":"de1befec952baa90620d5e02018428da3ef44ee4d913af7ec44077b7226325ef","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:16:49.987483Z","signature_b64":"7zyKcLl9JN9lt9YhVFzWQQ4fuTy5hXvZ35MOkkJhgsAzmIL768+V69VLtsddmp+ZlhfWyPvEY8LV5eunrxmsCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de1befec952baa90620d5e02018428da3ef44ee4d913af7ec44077b7226325ef","last_reissued_at":"2026-07-05T05:16:49.987131Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:16:49.987131Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2201.07821","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:16:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"D4hQkWgtoU8Z6N7UGqIfGxpR5KP8CiX3QeqX6vFRQjxu09Sk2z0D7R/GKwo+vNzvwLyrsk4X1UyCp5VQVTPDBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:29:46.290926Z"},"content_sha256":"ab01fa965e3b824a58953fee5058ec32984fff1152ea5954a774f0afec660ad0","schema_version":"1.0","event_id":"sha256:ab01fa965e3b824a58953fee5058ec32984fff1152ea5954a774f0afec660ad0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:3YN673EVFOVJAYQNLYBADBBI3I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Building a Performance Model for Deep Learning Recommendation Model Training on GPUs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.PF"],"primary_cat":"cs.LG","authors_text":"Arun Kejariwal, Changkyu Kim, Ehsan K. Ardestani, Jaewon Lee, John D. Owens, John Lundell, Louis Feng, Zhongyi Lin","submitted_at":"2022-01-19T19:05:42Z","abstract_excerpt":"We devise a performance model for GPU training of Deep Learning Recommendation Models (DLRM), whose GPU utilization is low compared to other well-optimized CV and NLP models. We show that both the device active time (the sum of kernel runtimes) but also the device idle time are important components of the overall device time. We therefore tackle them separately by (1) flexibly adopting heuristic-based and ML-based kernel performance models for operators that dominate the device active time, and (2) categorizing operator overheads into five types to determine quantitatively their contribution t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.07821","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/2201.07821/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":"2026-07-05T05:16:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SMDV5D2hmhaN6KopociMkcDljoQIGJdMhelbaQoct7jJOYH/oqBRKhyk6kxkC5N5C67AGpcG08nQK9vsbnX9Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:29:46.291300Z"},"content_sha256":"11cfbef5f2834a7bc464df9da8e8b1527f97bad4106d9d4da15079f312899536","schema_version":"1.0","event_id":"sha256:11cfbef5f2834a7bc464df9da8e8b1527f97bad4106d9d4da15079f312899536"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3YN673EVFOVJAYQNLYBADBBI3I/bundle.json","state_url":"https://pith.science/pith/3YN673EVFOVJAYQNLYBADBBI3I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3YN673EVFOVJAYQNLYBADBBI3I/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":"2026-07-07T08:29:46Z","links":{"resolver":"https://pith.science/pith/3YN673EVFOVJAYQNLYBADBBI3I","bundle":"https://pith.science/pith/3YN673EVFOVJAYQNLYBADBBI3I/bundle.json","state":"https://pith.science/pith/3YN673EVFOVJAYQNLYBADBBI3I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3YN673EVFOVJAYQNLYBADBBI3I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:3YN673EVFOVJAYQNLYBADBBI3I","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":"64b5d9c68555d0ccf16c8694a29931dbb4aacff15629b3c76a9b12c768d4d846","cross_cats_sorted":["cs.PF"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-01-19T19:05:42Z","title_canon_sha256":"f5bc5ef2532bd9243ab9d2d79e69b40351de8d876652fe2f17c0db919cfcf1cc"},"schema_version":"1.0","source":{"id":"2201.07821","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.07821","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"arxiv_version","alias_value":"2201.07821v2","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.07821","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"pith_short_12","alias_value":"3YN673EVFOVJ","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"pith_short_16","alias_value":"3YN673EVFOVJAYQN","created_at":"2026-07-05T05:16:49Z"},{"alias_kind":"pith_short_8","alias_value":"3YN673EV","created_at":"2026-07-05T05:16:49Z"}],"graph_snapshots":[{"event_id":"sha256:11cfbef5f2834a7bc464df9da8e8b1527f97bad4106d9d4da15079f312899536","target":"graph","created_at":"2026-07-05T05:16:49Z","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/2201.07821/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We devise a performance model for GPU training of Deep Learning Recommendation Models (DLRM), whose GPU utilization is low compared to other well-optimized CV and NLP models. We show that both the device active time (the sum of kernel runtimes) but also the device idle time are important components of the overall device time. We therefore tackle them separately by (1) flexibly adopting heuristic-based and ML-based kernel performance models for operators that dominate the device active time, and (2) categorizing operator overheads into five types to determine quantitatively their contribution t","authors_text":"Arun Kejariwal, Changkyu Kim, Ehsan K. Ardestani, Jaewon Lee, John D. Owens, John Lundell, Louis Feng, Zhongyi Lin","cross_cats":["cs.PF"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-01-19T19:05:42Z","title":"Building a Performance Model for Deep Learning Recommendation Model Training on GPUs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.07821","kind":"arxiv","version":2},"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:ab01fa965e3b824a58953fee5058ec32984fff1152ea5954a774f0afec660ad0","target":"record","created_at":"2026-07-05T05:16:49Z","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":"64b5d9c68555d0ccf16c8694a29931dbb4aacff15629b3c76a9b12c768d4d846","cross_cats_sorted":["cs.PF"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-01-19T19:05:42Z","title_canon_sha256":"f5bc5ef2532bd9243ab9d2d79e69b40351de8d876652fe2f17c0db919cfcf1cc"},"schema_version":"1.0","source":{"id":"2201.07821","kind":"arxiv","version":2}},"canonical_sha256":"de1befec952baa90620d5e02018428da3ef44ee4d913af7ec44077b7226325ef","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"de1befec952baa90620d5e02018428da3ef44ee4d913af7ec44077b7226325ef","first_computed_at":"2026-07-05T05:16:49.987131Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:16:49.987131Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7zyKcLl9JN9lt9YhVFzWQQ4fuTy5hXvZ35MOkkJhgsAzmIL768+V69VLtsddmp+ZlhfWyPvEY8LV5eunrxmsCw==","signature_status":"signed_v1","signed_at":"2026-07-05T05:16:49.987483Z","signed_message":"canonical_sha256_bytes"},"source_id":"2201.07821","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ab01fa965e3b824a58953fee5058ec32984fff1152ea5954a774f0afec660ad0","sha256:11cfbef5f2834a7bc464df9da8e8b1527f97bad4106d9d4da15079f312899536"],"state_sha256":"7345eda78071483f07a0bc70dd45937ecb43924f62335a94d7cd788bf32b5f4f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A/bnrt5rFdHyypCGWiFgf+JP4COLXxL4M8JSM948f+WlYwxLt613xs69X8ZcJlCEvZt26pr9QFsYyKGeNiwvDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T08:29:46.293301Z","bundle_sha256":"feaec27097e39966042ec377c1a5b0ae25b88975794b5424bbd63a939f8a2470"}}