{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:4T7LATV7TTH4FISCWRIIWETBPD","short_pith_number":"pith:4T7LATV7","schema_version":"1.0","canonical_sha256":"e4feb04ebf9ccfc2a242b4508b126178c8cfb443a01c7ea560d156b0bdbe3956","source":{"kind":"arxiv","id":"2008.09213","version":1},"attestation_state":"computed","paper":{"title":"Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Amar Phanishayee, Deepak Narayanan, Fiodar Kazhamiaka, Keshav Santhanam, Matei Zaharia","submitted_at":"2020-08-20T21:25:31Z","abstract_excerpt":"Specialized accelerators such as GPUs, TPUs, FPGAs, and custom ASICs have been increasingly deployed to train deep learning models. These accelerators exhibit heterogeneous performance behavior across model architectures. Existing schedulers for clusters of accelerators, which are used to arbitrate these expensive training resources across many users, have shown how to optimize for various multi-job, multi-user objectives, like fairness and makespan. Unfortunately, existing schedulers largely do not consider performance heterogeneity. In this paper, we propose Gavel, a heterogeneity-aware sche"},"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":"2008.09213","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2020-08-20T21:25:31Z","cross_cats_sorted":[],"title_canon_sha256":"bd7a43e8ca9dc40cda786ed45f53bd499d29a6b2752c847c97eb95c256c89fea","abstract_canon_sha256":"3910ec4b18bc964eb64bb7488fc26d5d2648f0c985c27e012c27a52b0302e416"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:28:41.983450Z","signature_b64":"ydBACmEbauM/v5PXT5bHjsDUP2Cto4J9fHQgDcGLr2JApDC3kEJdpoeDozqzDiUsoQW4rY1/kfEXq5V6wZ/1AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4feb04ebf9ccfc2a242b4508b126178c8cfb443a01c7ea560d156b0bdbe3956","last_reissued_at":"2026-07-05T01:28:41.982873Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:28:41.982873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Amar Phanishayee, Deepak Narayanan, Fiodar Kazhamiaka, Keshav Santhanam, Matei Zaharia","submitted_at":"2020-08-20T21:25:31Z","abstract_excerpt":"Specialized accelerators such as GPUs, TPUs, FPGAs, and custom ASICs have been increasingly deployed to train deep learning models. These accelerators exhibit heterogeneous performance behavior across model architectures. Existing schedulers for clusters of accelerators, which are used to arbitrate these expensive training resources across many users, have shown how to optimize for various multi-job, multi-user objectives, like fairness and makespan. Unfortunately, existing schedulers largely do not consider performance heterogeneity. In this paper, we propose Gavel, a heterogeneity-aware sche"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.09213","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/2008.09213/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":"2008.09213","created_at":"2026-07-05T01:28:41.982941+00:00"},{"alias_kind":"arxiv_version","alias_value":"2008.09213v1","created_at":"2026-07-05T01:28:41.982941+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.09213","created_at":"2026-07-05T01:28:41.982941+00:00"},{"alias_kind":"pith_short_12","alias_value":"4T7LATV7TTH4","created_at":"2026-07-05T01:28:41.982941+00:00"},{"alias_kind":"pith_short_16","alias_value":"4T7LATV7TTH4FISC","created_at":"2026-07-05T01:28:41.982941+00:00"},{"alias_kind":"pith_short_8","alias_value":"4T7LATV7","created_at":"2026-07-05T01:28:41.982941+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/4T7LATV7TTH4FISCWRIIWETBPD","json":"https://pith.science/pith/4T7LATV7TTH4FISCWRIIWETBPD.json","graph_json":"https://pith.science/api/pith-number/4T7LATV7TTH4FISCWRIIWETBPD/graph.json","events_json":"https://pith.science/api/pith-number/4T7LATV7TTH4FISCWRIIWETBPD/events.json","paper":"https://pith.science/paper/4T7LATV7"},"agent_actions":{"view_html":"https://pith.science/pith/4T7LATV7TTH4FISCWRIIWETBPD","download_json":"https://pith.science/pith/4T7LATV7TTH4FISCWRIIWETBPD.json","view_paper":"https://pith.science/paper/4T7LATV7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2008.09213&json=true","fetch_graph":"https://pith.science/api/pith-number/4T7LATV7TTH4FISCWRIIWETBPD/graph.json","fetch_events":"https://pith.science/api/pith-number/4T7LATV7TTH4FISCWRIIWETBPD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4T7LATV7TTH4FISCWRIIWETBPD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4T7LATV7TTH4FISCWRIIWETBPD/action/storage_attestation","attest_author":"https://pith.science/pith/4T7LATV7TTH4FISCWRIIWETBPD/action/author_attestation","sign_citation":"https://pith.science/pith/4T7LATV7TTH4FISCWRIIWETBPD/action/citation_signature","submit_replication":"https://pith.science/pith/4T7LATV7TTH4FISCWRIIWETBPD/action/replication_record"}},"created_at":"2026-07-05T01:28:41.982941+00:00","updated_at":"2026-07-05T01:28:41.982941+00:00"}