{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CHYS7TFTRMRLTDQLWR5C2YGVWG","short_pith_number":"pith:CHYS7TFT","schema_version":"1.0","canonical_sha256":"11f12fccb38b22b98e0bb47a2d60d5b194c113bcda3cc2bd0c1b94c91351ce66","source":{"kind":"arxiv","id":"2508.18556","version":1},"attestation_state":"computed","paper":{"title":"Managing Multi Instance GPUs for High Throughput and Energy Savings","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Abhijeet Saraha, Chris Porter, Santosh Pande, Yuanbo Li","submitted_at":"2025-08-25T23:15:49Z","abstract_excerpt":"Modern GPUs such as the Ampere series (A30, A100) as well as the Hopper series (H100, H200) offer performance as well as security isolation features. They also support a good amount of concurrency, but taking advantage of it can be quite challenging due to the complex constraints on partitioning the chip.\n  In this work, we develop partitioning and scheduling schemes for a variety of workloads, ranging from scientific to modern ML workloads, including LLMs. We develop several schemes involving dynamic memory estimation, partition fusion and partition fission. We also support process restart to"},"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":"2508.18556","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-08-25T23:15:49Z","cross_cats_sorted":[],"title_canon_sha256":"7812da3dada4b22014050b1b7b5d615e89c74b1c950c3373f5f0baca37cba338","abstract_canon_sha256":"148f34c4dd16cb9b4ff4626d5b4c5dd41974f73f60f8f33ae933478e189a7866"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:59:26.628871Z","signature_b64":"ySEa/yKnukBhlFBynN4VO7lKecvmqeG+ssHO/qD1+iPHCsTA6QXraQkqR42lOenGm3nMYVX0luTYr6eACKwODg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"11f12fccb38b22b98e0bb47a2d60d5b194c113bcda3cc2bd0c1b94c91351ce66","last_reissued_at":"2026-07-05T11:59:26.628408Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:59:26.628408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Managing Multi Instance GPUs for High Throughput and Energy Savings","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Abhijeet Saraha, Chris Porter, Santosh Pande, Yuanbo Li","submitted_at":"2025-08-25T23:15:49Z","abstract_excerpt":"Modern GPUs such as the Ampere series (A30, A100) as well as the Hopper series (H100, H200) offer performance as well as security isolation features. They also support a good amount of concurrency, but taking advantage of it can be quite challenging due to the complex constraints on partitioning the chip.\n  In this work, we develop partitioning and scheduling schemes for a variety of workloads, ranging from scientific to modern ML workloads, including LLMs. We develop several schemes involving dynamic memory estimation, partition fusion and partition fission. We also support process restart to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.18556","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/2508.18556/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":"2508.18556","created_at":"2026-07-05T11:59:26.628468+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.18556v1","created_at":"2026-07-05T11:59:26.628468+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.18556","created_at":"2026-07-05T11:59:26.628468+00:00"},{"alias_kind":"pith_short_12","alias_value":"CHYS7TFTRMRL","created_at":"2026-07-05T11:59:26.628468+00:00"},{"alias_kind":"pith_short_16","alias_value":"CHYS7TFTRMRLTDQL","created_at":"2026-07-05T11:59:26.628468+00:00"},{"alias_kind":"pith_short_8","alias_value":"CHYS7TFT","created_at":"2026-07-05T11:59:26.628468+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/CHYS7TFTRMRLTDQLWR5C2YGVWG","json":"https://pith.science/pith/CHYS7TFTRMRLTDQLWR5C2YGVWG.json","graph_json":"https://pith.science/api/pith-number/CHYS7TFTRMRLTDQLWR5C2YGVWG/graph.json","events_json":"https://pith.science/api/pith-number/CHYS7TFTRMRLTDQLWR5C2YGVWG/events.json","paper":"https://pith.science/paper/CHYS7TFT"},"agent_actions":{"view_html":"https://pith.science/pith/CHYS7TFTRMRLTDQLWR5C2YGVWG","download_json":"https://pith.science/pith/CHYS7TFTRMRLTDQLWR5C2YGVWG.json","view_paper":"https://pith.science/paper/CHYS7TFT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.18556&json=true","fetch_graph":"https://pith.science/api/pith-number/CHYS7TFTRMRLTDQLWR5C2YGVWG/graph.json","fetch_events":"https://pith.science/api/pith-number/CHYS7TFTRMRLTDQLWR5C2YGVWG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CHYS7TFTRMRLTDQLWR5C2YGVWG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CHYS7TFTRMRLTDQLWR5C2YGVWG/action/storage_attestation","attest_author":"https://pith.science/pith/CHYS7TFTRMRLTDQLWR5C2YGVWG/action/author_attestation","sign_citation":"https://pith.science/pith/CHYS7TFTRMRLTDQLWR5C2YGVWG/action/citation_signature","submit_replication":"https://pith.science/pith/CHYS7TFTRMRLTDQLWR5C2YGVWG/action/replication_record"}},"created_at":"2026-07-05T11:59:26.628468+00:00","updated_at":"2026-07-05T11:59:26.628468+00:00"}