{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:FX5YDHYV4EFNXXNOR7NKCESH7H","short_pith_number":"pith:FX5YDHYV","canonical_record":{"source":{"id":"2404.07947","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-15T06:21:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fa2f2d6abd90c6f61fc0b3061208b0ad8048fd3c2d2f87c97b6d10abff9ecfe6","abstract_canon_sha256":"14cb124158599b1baaab7c77ba404a511a1d7a6dc233945136792baa26b73f1b"},"schema_version":"1.0"},"canonical_sha256":"2dfb819f15e10adbddae8fdaa11247f9e4ce141ceb5501c465a3601a0418a6bc","source":{"kind":"arxiv","id":"2404.07947","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.07947","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"arxiv_version","alias_value":"2404.07947v1","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.07947","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"pith_short_12","alias_value":"FX5YDHYV4EFN","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"pith_short_16","alias_value":"FX5YDHYV4EFNXXNO","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"pith_short_8","alias_value":"FX5YDHYV","created_at":"2026-07-05T08:06:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:FX5YDHYV4EFNXXNOR7NKCESH7H","target":"record","payload":{"canonical_record":{"source":{"id":"2404.07947","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-15T06:21:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fa2f2d6abd90c6f61fc0b3061208b0ad8048fd3c2d2f87c97b6d10abff9ecfe6","abstract_canon_sha256":"14cb124158599b1baaab7c77ba404a511a1d7a6dc233945136792baa26b73f1b"},"schema_version":"1.0"},"canonical_sha256":"2dfb819f15e10adbddae8fdaa11247f9e4ce141ceb5501c465a3601a0418a6bc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:06:59.762473Z","signature_b64":"cM8PU42N6zxLTJf6iBib9uy2/zboryrcD8BNBaATjRETUjeKMnBDf2hGfv6P1Isy2MP/urmskd+CI1LttjhKDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2dfb819f15e10adbddae8fdaa11247f9e4ce141ceb5501c465a3601a0418a6bc","last_reissued_at":"2026-07-05T08:06:59.761966Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:06:59.761966Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.07947","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:06:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eUWq0fn8cDugkTF472kVynEd1qhhPNzC4Rf474ujAgSVEjL6cCdH4dTpyRYK3XlFm7D6xx9Vj8ZUMqyanROgAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T09:17:41.048234Z"},"content_sha256":"cee07076a70a73ed878ba9646e034f48c423b43ba4d35b68004fdb9f00afcede","schema_version":"1.0","event_id":"sha256:cee07076a70a73ed878ba9646e034f48c423b43ba4d35b68004fdb9f00afcede"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:FX5YDHYV4EFNXXNOR7NKCESH7H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ExeGPT: Constraint-Aware Resource Scheduling for LLM Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Du-Seong Chang, Hyungjun Oh, Jaemin Kim, Jiwon Seo, Junyeol Lee, Kihong Kim, Sungkyun Kim","submitted_at":"2024-03-15T06:21:56Z","abstract_excerpt":"This paper presents ExeGPT, a distributed system designed for constraint-aware LLM inference. ExeGPT finds and runs with an optimal execution schedule to maximize inference throughput while satisfying a given latency constraint. By leveraging the distribution of input and output sequences, it effectively allocates resources and determines optimal execution configurations, including batch sizes and partial tensor parallelism. We also introduce two scheduling strategies based on Round-Robin Allocation and Workload-Aware Allocation policies, suitable for different NLP workloads. We evaluate ExeGP"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.07947","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/2404.07947/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":"2026-07-05T08:06:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"siWllM3XEt4pygl21kxIgsqYtbbeLtd9loeg1xWKpq0D6+KuJUkF1FAcxSkMSNxIQRckjvWpB/qDrBYeV2vOAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T09:17:41.049186Z"},"content_sha256":"1cf631906e9e955603e4ef7753150f45333850ab50bc0efc2740548f0b40cea1","schema_version":"1.0","event_id":"sha256:1cf631906e9e955603e4ef7753150f45333850ab50bc0efc2740548f0b40cea1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FX5YDHYV4EFNXXNOR7NKCESH7H/bundle.json","state_url":"https://pith.science/pith/FX5YDHYV4EFNXXNOR7NKCESH7H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FX5YDHYV4EFNXXNOR7NKCESH7H/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":"2026-08-17T09:17:41Z","links":{"resolver":"https://pith.science/pith/FX5YDHYV4EFNXXNOR7NKCESH7H","bundle":"https://pith.science/pith/FX5YDHYV4EFNXXNOR7NKCESH7H/bundle.json","state":"https://pith.science/pith/FX5YDHYV4EFNXXNOR7NKCESH7H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FX5YDHYV4EFNXXNOR7NKCESH7H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:FX5YDHYV4EFNXXNOR7NKCESH7H","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":"14cb124158599b1baaab7c77ba404a511a1d7a6dc233945136792baa26b73f1b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-15T06:21:56Z","title_canon_sha256":"fa2f2d6abd90c6f61fc0b3061208b0ad8048fd3c2d2f87c97b6d10abff9ecfe6"},"schema_version":"1.0","source":{"id":"2404.07947","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.07947","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"arxiv_version","alias_value":"2404.07947v1","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.07947","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"pith_short_12","alias_value":"FX5YDHYV4EFN","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"pith_short_16","alias_value":"FX5YDHYV4EFNXXNO","created_at":"2026-07-05T08:06:59Z"},{"alias_kind":"pith_short_8","alias_value":"FX5YDHYV","created_at":"2026-07-05T08:06:59Z"}],"graph_snapshots":[{"event_id":"sha256:1cf631906e9e955603e4ef7753150f45333850ab50bc0efc2740548f0b40cea1","target":"graph","created_at":"2026-07-05T08:06:59Z","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/2404.07947/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper presents ExeGPT, a distributed system designed for constraint-aware LLM inference. ExeGPT finds and runs with an optimal execution schedule to maximize inference throughput while satisfying a given latency constraint. By leveraging the distribution of input and output sequences, it effectively allocates resources and determines optimal execution configurations, including batch sizes and partial tensor parallelism. We also introduce two scheduling strategies based on Round-Robin Allocation and Workload-Aware Allocation policies, suitable for different NLP workloads. We evaluate ExeGP","authors_text":"Du-Seong Chang, Hyungjun Oh, Jaemin Kim, Jiwon Seo, Junyeol Lee, Kihong Kim, Sungkyun Kim","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-15T06:21:56Z","title":"ExeGPT: Constraint-Aware Resource Scheduling for LLM Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.07947","kind":"arxiv","version":1},"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:cee07076a70a73ed878ba9646e034f48c423b43ba4d35b68004fdb9f00afcede","target":"record","created_at":"2026-07-05T08:06:59Z","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":"14cb124158599b1baaab7c77ba404a511a1d7a6dc233945136792baa26b73f1b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-15T06:21:56Z","title_canon_sha256":"fa2f2d6abd90c6f61fc0b3061208b0ad8048fd3c2d2f87c97b6d10abff9ecfe6"},"schema_version":"1.0","source":{"id":"2404.07947","kind":"arxiv","version":1}},"canonical_sha256":"2dfb819f15e10adbddae8fdaa11247f9e4ce141ceb5501c465a3601a0418a6bc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2dfb819f15e10adbddae8fdaa11247f9e4ce141ceb5501c465a3601a0418a6bc","first_computed_at":"2026-07-05T08:06:59.761966Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:06:59.761966Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cM8PU42N6zxLTJf6iBib9uy2/zboryrcD8BNBaATjRETUjeKMnBDf2hGfv6P1Isy2MP/urmskd+CI1LttjhKDA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:06:59.762473Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.07947","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cee07076a70a73ed878ba9646e034f48c423b43ba4d35b68004fdb9f00afcede","sha256:1cf631906e9e955603e4ef7753150f45333850ab50bc0efc2740548f0b40cea1"],"state_sha256":"ba6b24dd4673e6ec72d01f7948198bb07a6b35ffdb0a601e228e7e23f1d46ee1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vP4JlkNZZtK81ZrSILMvCYhZ5RhAHyJcNd1/AHhMkL7QQ6QBBnOAL5pexOeNpOOsGNEBxbo/pGwnBcs2qxYVDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T09:17:41.056549Z","bundle_sha256":"75d3accbcbb90a05de2c0b2826f3ad10055f20f87d6b9663a24ac2308092b2af"}}