{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:6YGEMVQPVQPMW7XTQHAHHMM5H2","short_pith_number":"pith:6YGEMVQP","schema_version":"1.0","canonical_sha256":"f60c46560fac1ecb7ef381c073b19d3e9871ec22a3952d40f506f79a345fd1d5","source":{"kind":"arxiv","id":"2311.03703","version":3},"attestation_state":"computed","paper":{"title":"Practical Performance Guarantees for Pipelined DNN Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Aaron Archer, Kuikui Liu, Matthew Fahrbach, Prakash Prabhu","submitted_at":"2023-11-07T03:55:39Z","abstract_excerpt":"We optimize pipeline parallelism for deep neural network (DNN) inference by partitioning model graphs into $k$ stages and minimizing the running time of the bottleneck stage, including communication. We give practical and effective algorithms for this NP-hard problem, but our emphasis is on tackling the practitioner's dilemma of deciding when a solution is good enough. To this end, we design novel mixed-integer programming (MIP) relaxations for proving lower bounds. Applying these methods to a diverse testbed of 369 production models, for $k \\in \\{2, 4, 8, 16, 32, 64\\}$, we empirically show th"},"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":"2311.03703","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-07T03:55:39Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"6d0f3a638af8da501fc642308e9dbff31d17b51cebef7007ff61bcfa168d38c3","abstract_canon_sha256":"25de8e0dd42026266f51ac5463569da2daac24b01a1fa02f1613ba1791ce2c23"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:00.035960Z","signature_b64":"260m5m5M4wUdfd6W+mlzKSgpqCRFX9pn8njGEgx0xoXwHjt7T9m68xuCbpfo/HX/fU3WS4nlO/IY0u2vLxVYCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f60c46560fac1ecb7ef381c073b19d3e9871ec22a3952d40f506f79a345fd1d5","last_reissued_at":"2026-07-05T08:27:00.035457Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:00.035457Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Practical Performance Guarantees for Pipelined DNN Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Aaron Archer, Kuikui Liu, Matthew Fahrbach, Prakash Prabhu","submitted_at":"2023-11-07T03:55:39Z","abstract_excerpt":"We optimize pipeline parallelism for deep neural network (DNN) inference by partitioning model graphs into $k$ stages and minimizing the running time of the bottleneck stage, including communication. We give practical and effective algorithms for this NP-hard problem, but our emphasis is on tackling the practitioner's dilemma of deciding when a solution is good enough. To this end, we design novel mixed-integer programming (MIP) relaxations for proving lower bounds. Applying these methods to a diverse testbed of 369 production models, for $k \\in \\{2, 4, 8, 16, 32, 64\\}$, we empirically show th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.03703","kind":"arxiv","version":3},"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/2311.03703/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":"2311.03703","created_at":"2026-07-05T08:27:00.035519+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.03703v3","created_at":"2026-07-05T08:27:00.035519+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.03703","created_at":"2026-07-05T08:27:00.035519+00:00"},{"alias_kind":"pith_short_12","alias_value":"6YGEMVQPVQPM","created_at":"2026-07-05T08:27:00.035519+00:00"},{"alias_kind":"pith_short_16","alias_value":"6YGEMVQPVQPMW7XT","created_at":"2026-07-05T08:27:00.035519+00:00"},{"alias_kind":"pith_short_8","alias_value":"6YGEMVQP","created_at":"2026-07-05T08:27:00.035519+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/6YGEMVQPVQPMW7XTQHAHHMM5H2","json":"https://pith.science/pith/6YGEMVQPVQPMW7XTQHAHHMM5H2.json","graph_json":"https://pith.science/api/pith-number/6YGEMVQPVQPMW7XTQHAHHMM5H2/graph.json","events_json":"https://pith.science/api/pith-number/6YGEMVQPVQPMW7XTQHAHHMM5H2/events.json","paper":"https://pith.science/paper/6YGEMVQP"},"agent_actions":{"view_html":"https://pith.science/pith/6YGEMVQPVQPMW7XTQHAHHMM5H2","download_json":"https://pith.science/pith/6YGEMVQPVQPMW7XTQHAHHMM5H2.json","view_paper":"https://pith.science/paper/6YGEMVQP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.03703&json=true","fetch_graph":"https://pith.science/api/pith-number/6YGEMVQPVQPMW7XTQHAHHMM5H2/graph.json","fetch_events":"https://pith.science/api/pith-number/6YGEMVQPVQPMW7XTQHAHHMM5H2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6YGEMVQPVQPMW7XTQHAHHMM5H2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6YGEMVQPVQPMW7XTQHAHHMM5H2/action/storage_attestation","attest_author":"https://pith.science/pith/6YGEMVQPVQPMW7XTQHAHHMM5H2/action/author_attestation","sign_citation":"https://pith.science/pith/6YGEMVQPVQPMW7XTQHAHHMM5H2/action/citation_signature","submit_replication":"https://pith.science/pith/6YGEMVQPVQPMW7XTQHAHHMM5H2/action/replication_record"}},"created_at":"2026-07-05T08:27:00.035519+00:00","updated_at":"2026-07-05T08:27:00.035519+00:00"}