{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4RW2SWHNCQZ3KTSVOLOITH2Y55","short_pith_number":"pith:4RW2SWHN","schema_version":"1.0","canonical_sha256":"e46da958ed1433b54e5572dc899f58ef59a7d351b60fc47dd8ae19218cf876ba","source":{"kind":"arxiv","id":"2109.01611","version":1},"attestation_state":"computed","paper":{"title":"Multi-model Machine Learning Inference Serving with GPU Spatial Partitioning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Jaehyuk Huh, Jongse Park, Seungbeom Choi, Sunho Lee, Yeonjae Kim, Youngjin Kwon","submitted_at":"2021-09-01T04:46:46Z","abstract_excerpt":"As machine learning techniques are applied to a widening range of applications, high throughput machine learning (ML) inference servers have become critical for online service applications. Such ML inference servers pose two challenges: first, they must provide a bounded latency for each request to support consistent service-level objective (SLO), and second, they can serve multiple heterogeneous ML models in a system as certain tasks involve invocation of multiple models and consolidating multiple models can improve system utilization. To address the two requirements of ML inference servers, "},"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":"2109.01611","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.DC","submitted_at":"2021-09-01T04:46:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f7a2f10ec1983de23869ca70fdd905c4bd25425fabb0f2409dbd8c55e967c63c","abstract_canon_sha256":"6edb2788ecacb4436d7d7460c9a3434e73819be5570da693f944f5ffa0003cc9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:11:20.123332Z","signature_b64":"JZBkqcYmkeyirUBXGVCIaVo22jBJsSi8BFI+X4jqCfAe0FnhlIomEIKoeRR+AJVqbRntVBLAYUS1m42+EtDQDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e46da958ed1433b54e5572dc899f58ef59a7d351b60fc47dd8ae19218cf876ba","last_reissued_at":"2026-07-05T03:11:20.122985Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:11:20.122985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-model Machine Learning Inference Serving with GPU Spatial Partitioning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Jaehyuk Huh, Jongse Park, Seungbeom Choi, Sunho Lee, Yeonjae Kim, Youngjin Kwon","submitted_at":"2021-09-01T04:46:46Z","abstract_excerpt":"As machine learning techniques are applied to a widening range of applications, high throughput machine learning (ML) inference servers have become critical for online service applications. Such ML inference servers pose two challenges: first, they must provide a bounded latency for each request to support consistent service-level objective (SLO), and second, they can serve multiple heterogeneous ML models in a system as certain tasks involve invocation of multiple models and consolidating multiple models can improve system utilization. To address the two requirements of ML inference servers, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.01611","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/2109.01611/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":"2109.01611","created_at":"2026-07-05T03:11:20.123045+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.01611v1","created_at":"2026-07-05T03:11:20.123045+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.01611","created_at":"2026-07-05T03:11:20.123045+00:00"},{"alias_kind":"pith_short_12","alias_value":"4RW2SWHNCQZ3","created_at":"2026-07-05T03:11:20.123045+00:00"},{"alias_kind":"pith_short_16","alias_value":"4RW2SWHNCQZ3KTSV","created_at":"2026-07-05T03:11:20.123045+00:00"},{"alias_kind":"pith_short_8","alias_value":"4RW2SWHN","created_at":"2026-07-05T03:11:20.123045+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.21276","citing_title":"LeMix: Unified Scheduling for LLM Training and Inference on Multi-GPU Systems","ref_index":7,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4RW2SWHNCQZ3KTSVOLOITH2Y55","json":"https://pith.science/pith/4RW2SWHNCQZ3KTSVOLOITH2Y55.json","graph_json":"https://pith.science/api/pith-number/4RW2SWHNCQZ3KTSVOLOITH2Y55/graph.json","events_json":"https://pith.science/api/pith-number/4RW2SWHNCQZ3KTSVOLOITH2Y55/events.json","paper":"https://pith.science/paper/4RW2SWHN"},"agent_actions":{"view_html":"https://pith.science/pith/4RW2SWHNCQZ3KTSVOLOITH2Y55","download_json":"https://pith.science/pith/4RW2SWHNCQZ3KTSVOLOITH2Y55.json","view_paper":"https://pith.science/paper/4RW2SWHN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.01611&json=true","fetch_graph":"https://pith.science/api/pith-number/4RW2SWHNCQZ3KTSVOLOITH2Y55/graph.json","fetch_events":"https://pith.science/api/pith-number/4RW2SWHNCQZ3KTSVOLOITH2Y55/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4RW2SWHNCQZ3KTSVOLOITH2Y55/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4RW2SWHNCQZ3KTSVOLOITH2Y55/action/storage_attestation","attest_author":"https://pith.science/pith/4RW2SWHNCQZ3KTSVOLOITH2Y55/action/author_attestation","sign_citation":"https://pith.science/pith/4RW2SWHNCQZ3KTSVOLOITH2Y55/action/citation_signature","submit_replication":"https://pith.science/pith/4RW2SWHNCQZ3KTSVOLOITH2Y55/action/replication_record"}},"created_at":"2026-07-05T03:11:20.123045+00:00","updated_at":"2026-07-05T03:11:20.123045+00:00"}