{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:USFPRWVH7HBYDP4EL4DVH2WK53","short_pith_number":"pith:USFPRWVH","schema_version":"1.0","canonical_sha256":"a48af8daa7f9c381bf845f0753eacaeefc87d163e1664d69f016141f7a7f2717","source":{"kind":"arxiv","id":"2607.29575","version":1},"attestation_state":"computed","paper":{"title":"SLIM: Saturation-Aware Lightweight Performance Modeling for LLM Serving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.PF"],"primary_cat":"cs.DC","authors_text":"Chen Wang, Ferran Agullo, Jordi Torres, Josep Ll. Berral, Pol G.Recasens, Yue Zhu","submitted_at":"2026-07-31T16:02:31Z","abstract_excerpt":"Large language model (LLM) serving commonly increases batch size to improve throughput, but performance eventually reaches a deployment-dependent plateau beyond which larger batches provide marginal gains while increasing latency and GPU memory consumption. Previous studies have attributed this behavior to HBM/DRAM bandwidth limitations, but the underlying causes have primarily been supported by conceptual arguments or high-level performance observations. As our first contribution, we present a detailed GPU characterization using hardware profiling techniques, demonstrating that throughput sat"},"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":"2607.29575","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-07-31T16:02:31Z","cross_cats_sorted":["cs.PF"],"title_canon_sha256":"d4090143408101eb8085eefb3bf51a2eb5c52370fb011a93fb8c7342cdba69a0","abstract_canon_sha256":"1283c3cc5c5c4a8e20ad868a2d05aee94caf711ba37b8416e85bda7583bbfa62"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-03T01:39:06.180267Z","signature_b64":"8OeeYu7aeYovXt/k0DdiUY21V4PmFlM8hIUme0GhKxyenjJ/Ui8yysWlo/ovTPyowBLxOydCIQ/E2xE9EnaZBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a48af8daa7f9c381bf845f0753eacaeefc87d163e1664d69f016141f7a7f2717","last_reissued_at":"2026-08-03T01:39:06.178514Z","signature_status":"signed_v1","first_computed_at":"2026-08-03T01:39:06.178514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SLIM: Saturation-Aware Lightweight Performance Modeling for LLM Serving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.PF"],"primary_cat":"cs.DC","authors_text":"Chen Wang, Ferran Agullo, Jordi Torres, Josep Ll. Berral, Pol G.Recasens, Yue Zhu","submitted_at":"2026-07-31T16:02:31Z","abstract_excerpt":"Large language model (LLM) serving commonly increases batch size to improve throughput, but performance eventually reaches a deployment-dependent plateau beyond which larger batches provide marginal gains while increasing latency and GPU memory consumption. Previous studies have attributed this behavior to HBM/DRAM bandwidth limitations, but the underlying causes have primarily been supported by conceptual arguments or high-level performance observations. As our first contribution, we present a detailed GPU characterization using hardware profiling techniques, demonstrating that throughput sat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.29575","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/2607.29575/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":"2607.29575","created_at":"2026-08-03T01:39:06.179558+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.29575v1","created_at":"2026-08-03T01:39:06.179558+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.29575","created_at":"2026-08-03T01:39:06.179558+00:00"},{"alias_kind":"pith_short_12","alias_value":"USFPRWVH7HBY","created_at":"2026-08-03T01:39:06.179558+00:00"},{"alias_kind":"pith_short_16","alias_value":"USFPRWVH7HBYDP4E","created_at":"2026-08-03T01:39:06.179558+00:00"},{"alias_kind":"pith_short_8","alias_value":"USFPRWVH","created_at":"2026-08-03T01:39:06.179558+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/USFPRWVH7HBYDP4EL4DVH2WK53","json":"https://pith.science/pith/USFPRWVH7HBYDP4EL4DVH2WK53.json","graph_json":"https://pith.science/api/pith-number/USFPRWVH7HBYDP4EL4DVH2WK53/graph.json","events_json":"https://pith.science/api/pith-number/USFPRWVH7HBYDP4EL4DVH2WK53/events.json","paper":"https://pith.science/paper/USFPRWVH"},"agent_actions":{"view_html":"https://pith.science/pith/USFPRWVH7HBYDP4EL4DVH2WK53","download_json":"https://pith.science/pith/USFPRWVH7HBYDP4EL4DVH2WK53.json","view_paper":"https://pith.science/paper/USFPRWVH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.29575&json=true","fetch_graph":"https://pith.science/api/pith-number/USFPRWVH7HBYDP4EL4DVH2WK53/graph.json","fetch_events":"https://pith.science/api/pith-number/USFPRWVH7HBYDP4EL4DVH2WK53/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/USFPRWVH7HBYDP4EL4DVH2WK53/action/timestamp_anchor","attest_storage":"https://pith.science/pith/USFPRWVH7HBYDP4EL4DVH2WK53/action/storage_attestation","attest_author":"https://pith.science/pith/USFPRWVH7HBYDP4EL4DVH2WK53/action/author_attestation","sign_citation":"https://pith.science/pith/USFPRWVH7HBYDP4EL4DVH2WK53/action/citation_signature","submit_replication":"https://pith.science/pith/USFPRWVH7HBYDP4EL4DVH2WK53/action/replication_record"}},"created_at":"2026-08-03T01:39:06.179558+00:00","updated_at":"2026-08-03T01:39:06.179558+00:00"}