{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:5CWDOJG3Z355BJNWVSGKOXPXWJ","short_pith_number":"pith:5CWDOJG3","canonical_record":{"source":{"id":"2407.00047","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-05T21:17:34Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"23171764e11d0d11cbaa6a5556fc3ca122a0673a7a55367a3ec77ab6c674787c","abstract_canon_sha256":"26c372d6dda5a7df0c8a3bb24bbaabf6ccd8a4e208ddce82ae8ee522fa833ba4"},"schema_version":"1.0"},"canonical_sha256":"e8ac3724dbcefbd0a5b6ac8ca75df7b26b547559158b2af64addb0f43e069b9f","source":{"kind":"arxiv","id":"2407.00047","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.00047","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"arxiv_version","alias_value":"2407.00047v2","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.00047","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"pith_short_12","alias_value":"5CWDOJG3Z355","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"pith_short_16","alias_value":"5CWDOJG3Z355BJNW","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"pith_short_8","alias_value":"5CWDOJG3","created_at":"2026-07-05T10:19:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:5CWDOJG3Z355BJNWVSGKOXPXWJ","target":"record","payload":{"canonical_record":{"source":{"id":"2407.00047","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-05T21:17:34Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"23171764e11d0d11cbaa6a5556fc3ca122a0673a7a55367a3ec77ab6c674787c","abstract_canon_sha256":"26c372d6dda5a7df0c8a3bb24bbaabf6ccd8a4e208ddce82ae8ee522fa833ba4"},"schema_version":"1.0"},"canonical_sha256":"e8ac3724dbcefbd0a5b6ac8ca75df7b26b547559158b2af64addb0f43e069b9f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:19:25.561452Z","signature_b64":"zlhImsjXmnlEyY9qYGUmhxx+9agXx1AdkbgCUJhtMDebeOC+bxuo4Bb4e7K4mLVyVz9Sw4wcenQnAEidWUXCBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e8ac3724dbcefbd0a5b6ac8ca75df7b26b547559158b2af64addb0f43e069b9f","last_reissued_at":"2026-07-05T10:19:25.560937Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:19:25.560937Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.00047","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:19:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BoubL5emlUaltU8KZD0+QVw+EcYpW/nPK//NnPBkalcoM+SAbm7ZbH62TPssu5yqIit7UoxjHaDOMyiRZzllAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T11:45:05.593121Z"},"content_sha256":"feeb737e6184586f63bc0e70a95b6bfc8c513fb7a1c1df1a7afed5a720cac320","schema_version":"1.0","event_id":"sha256:feeb737e6184586f63bc0e70a95b6bfc8c513fb7a1c1df1a7afed5a720cac320"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:5CWDOJG3Z355BJNWVSGKOXPXWJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Queue management for slo-oriented large language model serving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.DC","authors_text":"Archit Patke, Chandra Narayanaswami, Christian Pinto, Dhemath Reddy, Haoran Qiu, Ravishankar Iyer, Saurabh Jha, Zbigniew Kalbarczyk","submitted_at":"2024-06-05T21:17:34Z","abstract_excerpt":"Large language model (LLM) serving is becoming an increasingly critical workload for cloud providers. Existing LLM serving systems focus on interactive requests, such as chatbots and coding assistants, with tight latency SLO requirements. However, when such systems execute batch requests that have relaxed SLOs along with interactive requests, it leads to poor multiplexing and inefficient resource utilization. To address these challenges, we propose QLM, a queue management system for LLM serving. QLM maintains batch and interactive requests across different models and SLOs in a request queue. O"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.00047","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/2407.00047/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:19:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Nv3PTnDkO2LYSp0kuSkbgCmYJHUBhKHrxl2Sj542dlZnB6R2gI41fqT4MTGpeGLt2uHvvFLYLSAwOqOsDMZADA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T11:45:05.594213Z"},"content_sha256":"26402a4bdf6e8c297a6e1af6a16b20391c4f25d167ef2ca0be9a0e2eff204f7c","schema_version":"1.0","event_id":"sha256:26402a4bdf6e8c297a6e1af6a16b20391c4f25d167ef2ca0be9a0e2eff204f7c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5CWDOJG3Z355BJNWVSGKOXPXWJ/bundle.json","state_url":"https://pith.science/pith/5CWDOJG3Z355BJNWVSGKOXPXWJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5CWDOJG3Z355BJNWVSGKOXPXWJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-11T11:45:05Z","links":{"resolver":"https://pith.science/pith/5CWDOJG3Z355BJNWVSGKOXPXWJ","bundle":"https://pith.science/pith/5CWDOJG3Z355BJNWVSGKOXPXWJ/bundle.json","state":"https://pith.science/pith/5CWDOJG3Z355BJNWVSGKOXPXWJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5CWDOJG3Z355BJNWVSGKOXPXWJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:5CWDOJG3Z355BJNWVSGKOXPXWJ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"26c372d6dda5a7df0c8a3bb24bbaabf6ccd8a4e208ddce82ae8ee522fa833ba4","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-05T21:17:34Z","title_canon_sha256":"23171764e11d0d11cbaa6a5556fc3ca122a0673a7a55367a3ec77ab6c674787c"},"schema_version":"1.0","source":{"id":"2407.00047","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.00047","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"arxiv_version","alias_value":"2407.00047v2","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.00047","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"pith_short_12","alias_value":"5CWDOJG3Z355","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"pith_short_16","alias_value":"5CWDOJG3Z355BJNW","created_at":"2026-07-05T10:19:25Z"},{"alias_kind":"pith_short_8","alias_value":"5CWDOJG3","created_at":"2026-07-05T10:19:25Z"}],"graph_snapshots":[{"event_id":"sha256:26402a4bdf6e8c297a6e1af6a16b20391c4f25d167ef2ca0be9a0e2eff204f7c","target":"graph","created_at":"2026-07-05T10:19:25Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2407.00047/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language model (LLM) serving is becoming an increasingly critical workload for cloud providers. Existing LLM serving systems focus on interactive requests, such as chatbots and coding assistants, with tight latency SLO requirements. However, when such systems execute batch requests that have relaxed SLOs along with interactive requests, it leads to poor multiplexing and inefficient resource utilization. To address these challenges, we propose QLM, a queue management system for LLM serving. QLM maintains batch and interactive requests across different models and SLOs in a request queue. O","authors_text":"Archit Patke, Chandra Narayanaswami, Christian Pinto, Dhemath Reddy, Haoran Qiu, Ravishankar Iyer, Saurabh Jha, Zbigniew Kalbarczyk","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-05T21:17:34Z","title":"Queue management for slo-oriented large language model serving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.00047","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:feeb737e6184586f63bc0e70a95b6bfc8c513fb7a1c1df1a7afed5a720cac320","target":"record","created_at":"2026-07-05T10:19:25Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"26c372d6dda5a7df0c8a3bb24bbaabf6ccd8a4e208ddce82ae8ee522fa833ba4","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-05T21:17:34Z","title_canon_sha256":"23171764e11d0d11cbaa6a5556fc3ca122a0673a7a55367a3ec77ab6c674787c"},"schema_version":"1.0","source":{"id":"2407.00047","kind":"arxiv","version":2}},"canonical_sha256":"e8ac3724dbcefbd0a5b6ac8ca75df7b26b547559158b2af64addb0f43e069b9f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e8ac3724dbcefbd0a5b6ac8ca75df7b26b547559158b2af64addb0f43e069b9f","first_computed_at":"2026-07-05T10:19:25.560937Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:19:25.560937Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zlhImsjXmnlEyY9qYGUmhxx+9agXx1AdkbgCUJhtMDebeOC+bxuo4Bb4e7K4mLVyVz9Sw4wcenQnAEidWUXCBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:19:25.561452Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.00047","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:feeb737e6184586f63bc0e70a95b6bfc8c513fb7a1c1df1a7afed5a720cac320","sha256:26402a4bdf6e8c297a6e1af6a16b20391c4f25d167ef2ca0be9a0e2eff204f7c"],"state_sha256":"4208d32dd7a553a36cc98c87b72aa013ca3921acaf0fa2dcbac09a53fa064131"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1pFfyKmiAMaw/ij9NNM2DGKjN9Vdu6M46PkuxfI8X1U5vDWqWQpIOwIuHnbI7UCprkQStvriBUmUHiOFWuu8AQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T11:45:05.602802Z","bundle_sha256":"826e992a56f9ba24cd07ee9271a9ecc8cadf79aff3143eca0be5190b3add2eaa"}}