{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:5J7YF5LWYBM3OR26YLHKOAW65X","short_pith_number":"pith:5J7YF5LW","schema_version":"1.0","canonical_sha256":"ea7f82f576c059b7475ec2cea702deede60b543fda9bc71f7e6f269cf355f2c3","source":{"kind":"arxiv","id":"2210.16691","version":2},"attestation_state":"computed","paper":{"title":"ALCOP: Automatic Load-Compute Pipelining in Deep Learning Compiler for AI-GPUs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Bei Yu, Guyue Huang, Liu Liu, Yang Bai, Yuan Xie, Yufei Ding, Yuke Wang","submitted_at":"2022-10-29T20:47:27Z","abstract_excerpt":"Pipelining between data loading and computation is a critical tensor program optimization for GPUs. In order to unleash the high performance of latest GPUs, we must perform a synergetic optimization of multi-stage pipelining across the multi-level buffer hierarchy of GPU. Existing frameworks rely on hand-written libraries such as cuBLAS to perform pipelining optimization, which is inextensible to new operators and un-composable with prior tensor compiler optimizations. This paper presents ALCOP, the first framework that is compiler-native and fully supports multi-stage multi-level pipelining. "},"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":"2210.16691","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2022-10-29T20:47:27Z","cross_cats_sorted":[],"title_canon_sha256":"58bcacc544036fd214bfae826398116859a75c94ef5e99468c6e17e9efb2752f","abstract_canon_sha256":"e1b07b458467b732a4538a595d5c4575ed3345a2f84a538205187e16edf92c69"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:07:33.871220Z","signature_b64":"KlARPMfWC12ZkuZr81doPDsR9nA9mbbu8CDod7zLOyHDtC56SDz9fSHFsksgTr/0MI3mIiLd6vTwdr8AXAucDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ea7f82f576c059b7475ec2cea702deede60b543fda9bc71f7e6f269cf355f2c3","last_reissued_at":"2026-07-05T06:07:33.870772Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:07:33.870772Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ALCOP: Automatic Load-Compute Pipelining in Deep Learning Compiler for AI-GPUs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Bei Yu, Guyue Huang, Liu Liu, Yang Bai, Yuan Xie, Yufei Ding, Yuke Wang","submitted_at":"2022-10-29T20:47:27Z","abstract_excerpt":"Pipelining between data loading and computation is a critical tensor program optimization for GPUs. In order to unleash the high performance of latest GPUs, we must perform a synergetic optimization of multi-stage pipelining across the multi-level buffer hierarchy of GPU. Existing frameworks rely on hand-written libraries such as cuBLAS to perform pipelining optimization, which is inextensible to new operators and un-composable with prior tensor compiler optimizations. This paper presents ALCOP, the first framework that is compiler-native and fully supports multi-stage multi-level pipelining. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.16691","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/2210.16691/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":"2210.16691","created_at":"2026-07-05T06:07:33.870829+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.16691v2","created_at":"2026-07-05T06:07:33.870829+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.16691","created_at":"2026-07-05T06:07:33.870829+00:00"},{"alias_kind":"pith_short_12","alias_value":"5J7YF5LWYBM3","created_at":"2026-07-05T06:07:33.870829+00:00"},{"alias_kind":"pith_short_16","alias_value":"5J7YF5LWYBM3OR26","created_at":"2026-07-05T06:07:33.870829+00:00"},{"alias_kind":"pith_short_8","alias_value":"5J7YF5LW","created_at":"2026-07-05T06:07:33.870829+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.21661","citing_title":"KPerfIR: Towards an Open and Compiler-centric Ecosystem for GPU Kernel Performance Tooling on Modern AI Workloads","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5J7YF5LWYBM3OR26YLHKOAW65X","json":"https://pith.science/pith/5J7YF5LWYBM3OR26YLHKOAW65X.json","graph_json":"https://pith.science/api/pith-number/5J7YF5LWYBM3OR26YLHKOAW65X/graph.json","events_json":"https://pith.science/api/pith-number/5J7YF5LWYBM3OR26YLHKOAW65X/events.json","paper":"https://pith.science/paper/5J7YF5LW"},"agent_actions":{"view_html":"https://pith.science/pith/5J7YF5LWYBM3OR26YLHKOAW65X","download_json":"https://pith.science/pith/5J7YF5LWYBM3OR26YLHKOAW65X.json","view_paper":"https://pith.science/paper/5J7YF5LW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.16691&json=true","fetch_graph":"https://pith.science/api/pith-number/5J7YF5LWYBM3OR26YLHKOAW65X/graph.json","fetch_events":"https://pith.science/api/pith-number/5J7YF5LWYBM3OR26YLHKOAW65X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5J7YF5LWYBM3OR26YLHKOAW65X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5J7YF5LWYBM3OR26YLHKOAW65X/action/storage_attestation","attest_author":"https://pith.science/pith/5J7YF5LWYBM3OR26YLHKOAW65X/action/author_attestation","sign_citation":"https://pith.science/pith/5J7YF5LWYBM3OR26YLHKOAW65X/action/citation_signature","submit_replication":"https://pith.science/pith/5J7YF5LWYBM3OR26YLHKOAW65X/action/replication_record"}},"created_at":"2026-07-05T06:07:33.870829+00:00","updated_at":"2026-07-05T06:07:33.870829+00:00"}