{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:C5B5QJMQJHGWYP7F3AYR3256DP","short_pith_number":"pith:C5B5QJMQ","canonical_record":{"source":{"id":"2412.14335","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2024-12-18T21:09:08Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"0cd0b332862d65541a19bf1f816b51d496cc1f83c03a1067737be1dbc07408e2","abstract_canon_sha256":"9ae2a48c83c4c22d297d499ed57f79ecc820b95c912083a036999760c23879a6"},"schema_version":"1.0"},"canonical_sha256":"1743d8259049cd6c3fe5d8311debbe1bcd2c15e9020772b24f4a8f61f646e65d","source":{"kind":"arxiv","id":"2412.14335","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.14335","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"arxiv_version","alias_value":"2412.14335v2","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.14335","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"pith_short_12","alias_value":"C5B5QJMQJHGW","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"pith_short_16","alias_value":"C5B5QJMQJHGWYP7F","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"pith_short_8","alias_value":"C5B5QJMQ","created_at":"2026-07-05T10:53:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:C5B5QJMQJHGWYP7F3AYR3256DP","target":"record","payload":{"canonical_record":{"source":{"id":"2412.14335","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2024-12-18T21:09:08Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"0cd0b332862d65541a19bf1f816b51d496cc1f83c03a1067737be1dbc07408e2","abstract_canon_sha256":"9ae2a48c83c4c22d297d499ed57f79ecc820b95c912083a036999760c23879a6"},"schema_version":"1.0"},"canonical_sha256":"1743d8259049cd6c3fe5d8311debbe1bcd2c15e9020772b24f4a8f61f646e65d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:53:50.250840Z","signature_b64":"jWxcnu9j+hzwhFf2pa4fnhB74FJ6fIG29mlxgRUf5FByILZs+BNZlhxh9ppqhNubbsF9ZL3p3gQBxdi1zrqiBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1743d8259049cd6c3fe5d8311debbe1bcd2c15e9020772b24f4a8f61f646e65d","last_reissued_at":"2026-07-05T10:53:50.250253Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:53:50.250253Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.14335","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":"2026-07-05T10:53:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"13W2APLLkfz9Z2Q2HNDkF/bTzZoOXLmnWPGQuI9rPt1WqDe1at8NYt1QK93dn8jC5zFAuIyD5GqvM0xlaxrpBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T09:34:38.448669Z"},"content_sha256":"59771fbb91c0a51a54e803df34e7d1d542eb8b7559835cb5874dc8759557f687","schema_version":"1.0","event_id":"sha256:59771fbb91c0a51a54e803df34e7d1d542eb8b7559835cb5874dc8759557f687"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:C5B5QJMQJHGWYP7F3AYR3256DP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimizing ML Concurrent Computation and Communication with GPU DMA Engines","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.AR","authors_text":"Anirudha Agrawal, Mahzabeen Islam, Shaizeen Aga, Suchita Pati","submitted_at":"2024-12-18T21:09:08Z","abstract_excerpt":"Concurrent computation and communication (C3) is a pervasive paradigm in ML and other domains, making its performance optimization crucial. In this paper, we carefully characterize C3 in ML on GPUs, which are most widely deployed for ML training and inference. We observe that while C3 leads to performance uplifts, the uplifts are far lower than ideal speedups (serial computation and communication versus maximum of computation or communication; all times from isolated executions). That is, C3 on average achieves only 21% of ideal speedup. This is so, due to known challenges of compute and memor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.14335","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/2412.14335/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":"2026-07-05T10:53:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZFwZgLUfvCpZHNs2TOigqx2Ly1yIHwlbQrMgypYcgGH3HSnzT858EyUvcIwujxr4yCLQBKkq8dbIyLQMCkozDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T09:34:38.449191Z"},"content_sha256":"0b14270731bddde5dc8977d51af471a2d62ace1afba408ac98ab72c1b44a00d1","schema_version":"1.0","event_id":"sha256:0b14270731bddde5dc8977d51af471a2d62ace1afba408ac98ab72c1b44a00d1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/C5B5QJMQJHGWYP7F3AYR3256DP/bundle.json","state_url":"https://pith.science/pith/C5B5QJMQJHGWYP7F3AYR3256DP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/C5B5QJMQJHGWYP7F3AYR3256DP/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":"2026-08-17T09:34:38Z","links":{"resolver":"https://pith.science/pith/C5B5QJMQJHGWYP7F3AYR3256DP","bundle":"https://pith.science/pith/C5B5QJMQJHGWYP7F3AYR3256DP/bundle.json","state":"https://pith.science/pith/C5B5QJMQJHGWYP7F3AYR3256DP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/C5B5QJMQJHGWYP7F3AYR3256DP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:C5B5QJMQJHGWYP7F3AYR3256DP","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":"9ae2a48c83c4c22d297d499ed57f79ecc820b95c912083a036999760c23879a6","cross_cats_sorted":["cs.DC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2024-12-18T21:09:08Z","title_canon_sha256":"0cd0b332862d65541a19bf1f816b51d496cc1f83c03a1067737be1dbc07408e2"},"schema_version":"1.0","source":{"id":"2412.14335","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.14335","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"arxiv_version","alias_value":"2412.14335v2","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.14335","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"pith_short_12","alias_value":"C5B5QJMQJHGW","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"pith_short_16","alias_value":"C5B5QJMQJHGWYP7F","created_at":"2026-07-05T10:53:50Z"},{"alias_kind":"pith_short_8","alias_value":"C5B5QJMQ","created_at":"2026-07-05T10:53:50Z"}],"graph_snapshots":[{"event_id":"sha256:0b14270731bddde5dc8977d51af471a2d62ace1afba408ac98ab72c1b44a00d1","target":"graph","created_at":"2026-07-05T10:53:50Z","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/2412.14335/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Concurrent computation and communication (C3) is a pervasive paradigm in ML and other domains, making its performance optimization crucial. In this paper, we carefully characterize C3 in ML on GPUs, which are most widely deployed for ML training and inference. We observe that while C3 leads to performance uplifts, the uplifts are far lower than ideal speedups (serial computation and communication versus maximum of computation or communication; all times from isolated executions). That is, C3 on average achieves only 21% of ideal speedup. This is so, due to known challenges of compute and memor","authors_text":"Anirudha Agrawal, Mahzabeen Islam, Shaizeen Aga, Suchita Pati","cross_cats":["cs.DC"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2024-12-18T21:09:08Z","title":"Optimizing ML Concurrent Computation and Communication with GPU DMA Engines"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.14335","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":"sha256:59771fbb91c0a51a54e803df34e7d1d542eb8b7559835cb5874dc8759557f687","target":"record","created_at":"2026-07-05T10:53:50Z","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":"9ae2a48c83c4c22d297d499ed57f79ecc820b95c912083a036999760c23879a6","cross_cats_sorted":["cs.DC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2024-12-18T21:09:08Z","title_canon_sha256":"0cd0b332862d65541a19bf1f816b51d496cc1f83c03a1067737be1dbc07408e2"},"schema_version":"1.0","source":{"id":"2412.14335","kind":"arxiv","version":2}},"canonical_sha256":"1743d8259049cd6c3fe5d8311debbe1bcd2c15e9020772b24f4a8f61f646e65d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1743d8259049cd6c3fe5d8311debbe1bcd2c15e9020772b24f4a8f61f646e65d","first_computed_at":"2026-07-05T10:53:50.250253Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:53:50.250253Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jWxcnu9j+hzwhFf2pa4fnhB74FJ6fIG29mlxgRUf5FByILZs+BNZlhxh9ppqhNubbsF9ZL3p3gQBxdi1zrqiBg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:53:50.250840Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.14335","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:59771fbb91c0a51a54e803df34e7d1d542eb8b7559835cb5874dc8759557f687","sha256:0b14270731bddde5dc8977d51af471a2d62ace1afba408ac98ab72c1b44a00d1"],"state_sha256":"ba76bfb5576a09261faee152454a8f9b8d8785f891a8414374363d8b561b269d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Jk99vwHZAaruw8H0TkUVtKfwA7PAQ1Rzl8g6Kwe37etPT/fcqodzuYG2DThVdD2fIGtKJ80re45PbWcyKryaCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T09:34:38.454194Z","bundle_sha256":"204c308e691b3f95f732b2da6ac32d2ee0d7686120e5994c82c081b8c4ce760f"}}