{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:IRIDTZ3VDEB44EW3C7HWVKVRGU","short_pith_number":"pith:IRIDTZ3V","canonical_record":{"source":{"id":"2410.17840","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-23T13:05:46Z","cross_cats_sorted":[],"title_canon_sha256":"427e63ebc7249b3bc69efe3fbf9bfec46467948c9928e6663e44e7b515b85b10","abstract_canon_sha256":"fab4154fba9b03c6a6d9b4df528d04259737e9a156964edaa6f16bce028290f7"},"schema_version":"1.0"},"canonical_sha256":"445039e7751903ce12db17cf6aaab1352aff3f68f728f1194782b958b0129b8d","source":{"kind":"arxiv","id":"2410.17840","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.17840","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"arxiv_version","alias_value":"2410.17840v2","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.17840","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"pith_short_12","alias_value":"IRIDTZ3VDEB4","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"pith_short_16","alias_value":"IRIDTZ3VDEB44EW3","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"pith_short_8","alias_value":"IRIDTZ3V","created_at":"2026-07-05T10:06:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:IRIDTZ3VDEB44EW3C7HWVKVRGU","target":"record","payload":{"canonical_record":{"source":{"id":"2410.17840","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-23T13:05:46Z","cross_cats_sorted":[],"title_canon_sha256":"427e63ebc7249b3bc69efe3fbf9bfec46467948c9928e6663e44e7b515b85b10","abstract_canon_sha256":"fab4154fba9b03c6a6d9b4df528d04259737e9a156964edaa6f16bce028290f7"},"schema_version":"1.0"},"canonical_sha256":"445039e7751903ce12db17cf6aaab1352aff3f68f728f1194782b958b0129b8d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:06:08.735064Z","signature_b64":"4GQg2qrbf5KCi4ifg3Uzd7lhlLfZ8PlIMAXviZja3+lApM+bLzmr0KW5Rvw2UJMyhHyaI/gVCZ5CKjF2b+cmAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"445039e7751903ce12db17cf6aaab1352aff3f68f728f1194782b958b0129b8d","last_reissued_at":"2026-07-05T10:06:08.734548Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:06:08.734548Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.17840","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":"2026-07-05T10:06:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gedbrUtorTilhFc9cufnifGdclxIU3zUudgpba8IXr8/3VikRYA+HH2iYmmFMlzeOQyDYR757ZQJuLISa86FCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T22:34:09.353997Z"},"content_sha256":"a142a2d0e135429ab8c83c35b2d9da6dd8562f2f2431ac7aa3cc061fbe61b8a8","schema_version":"1.0","event_id":"sha256:a142a2d0e135429ab8c83c35b2d9da6dd8562f2f2431ac7aa3cc061fbe61b8a8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:IRIDTZ3VDEB44EW3C7HWVKVRGU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Is the GPU Half-Empty or Half-Full? Practical Scheduling Techniques for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bruce Fontaine, Daya Khudia, Ferdi Kossmann, Michael Cafarella, Samuel Madden","submitted_at":"2024-10-23T13:05:46Z","abstract_excerpt":"Serving systems for Large Language Models (LLMs) improve throughput by processing several requests concurrently. However, multiplexing hardware resources between concurrent requests involves non-trivial scheduling decisions. Practical serving systems typically implement these decisions at two levels: First, a load balancer routes requests to different servers which each hold a replica of the LLM. Then, on each server, an engine-level scheduler decides when to run a request, or when to queue or preempt it. Improved scheduling policies may benefit a wide range of LLM deployments and can often be"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.17840","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/2410.17840/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":"2026-07-05T10:06:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7UtaDW6T123cot3Tl/tqk6fmbtTax3qPAjpeHNSvzS9NEVNUrh6IXfyVtPhI6mNQWCMLt8dbNup8PPI/pSmLAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T22:34:09.354706Z"},"content_sha256":"deb15dda4118b179d72f4786d342a219465b47553768eb40a479885ff0b4987c","schema_version":"1.0","event_id":"sha256:deb15dda4118b179d72f4786d342a219465b47553768eb40a479885ff0b4987c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IRIDTZ3VDEB44EW3C7HWVKVRGU/bundle.json","state_url":"https://pith.science/pith/IRIDTZ3VDEB44EW3C7HWVKVRGU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IRIDTZ3VDEB44EW3C7HWVKVRGU/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":"2026-08-03T22:34:09Z","links":{"resolver":"https://pith.science/pith/IRIDTZ3VDEB44EW3C7HWVKVRGU","bundle":"https://pith.science/pith/IRIDTZ3VDEB44EW3C7HWVKVRGU/bundle.json","state":"https://pith.science/pith/IRIDTZ3VDEB44EW3C7HWVKVRGU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IRIDTZ3VDEB44EW3C7HWVKVRGU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:IRIDTZ3VDEB44EW3C7HWVKVRGU","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":"fab4154fba9b03c6a6d9b4df528d04259737e9a156964edaa6f16bce028290f7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-23T13:05:46Z","title_canon_sha256":"427e63ebc7249b3bc69efe3fbf9bfec46467948c9928e6663e44e7b515b85b10"},"schema_version":"1.0","source":{"id":"2410.17840","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.17840","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"arxiv_version","alias_value":"2410.17840v2","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.17840","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"pith_short_12","alias_value":"IRIDTZ3VDEB4","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"pith_short_16","alias_value":"IRIDTZ3VDEB44EW3","created_at":"2026-07-05T10:06:08Z"},{"alias_kind":"pith_short_8","alias_value":"IRIDTZ3V","created_at":"2026-07-05T10:06:08Z"}],"graph_snapshots":[{"event_id":"sha256:deb15dda4118b179d72f4786d342a219465b47553768eb40a479885ff0b4987c","target":"graph","created_at":"2026-07-05T10:06:08Z","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/2410.17840/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Serving systems for Large Language Models (LLMs) improve throughput by processing several requests concurrently. However, multiplexing hardware resources between concurrent requests involves non-trivial scheduling decisions. Practical serving systems typically implement these decisions at two levels: First, a load balancer routes requests to different servers which each hold a replica of the LLM. Then, on each server, an engine-level scheduler decides when to run a request, or when to queue or preempt it. Improved scheduling policies may benefit a wide range of LLM deployments and can often be","authors_text":"Bruce Fontaine, Daya Khudia, Ferdi Kossmann, Michael Cafarella, Samuel Madden","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-23T13:05:46Z","title":"Is the GPU Half-Empty or Half-Full? Practical Scheduling Techniques for LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.17840","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:a142a2d0e135429ab8c83c35b2d9da6dd8562f2f2431ac7aa3cc061fbe61b8a8","target":"record","created_at":"2026-07-05T10:06:08Z","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":"fab4154fba9b03c6a6d9b4df528d04259737e9a156964edaa6f16bce028290f7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-23T13:05:46Z","title_canon_sha256":"427e63ebc7249b3bc69efe3fbf9bfec46467948c9928e6663e44e7b515b85b10"},"schema_version":"1.0","source":{"id":"2410.17840","kind":"arxiv","version":2}},"canonical_sha256":"445039e7751903ce12db17cf6aaab1352aff3f68f728f1194782b958b0129b8d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"445039e7751903ce12db17cf6aaab1352aff3f68f728f1194782b958b0129b8d","first_computed_at":"2026-07-05T10:06:08.734548Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:06:08.734548Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4GQg2qrbf5KCi4ifg3Uzd7lhlLfZ8PlIMAXviZja3+lApM+bLzmr0KW5Rvw2UJMyhHyaI/gVCZ5CKjF2b+cmAw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:06:08.735064Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.17840","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a142a2d0e135429ab8c83c35b2d9da6dd8562f2f2431ac7aa3cc061fbe61b8a8","sha256:deb15dda4118b179d72f4786d342a219465b47553768eb40a479885ff0b4987c"],"state_sha256":"b5e7c0334d6ac3b525ad66a29b19c7338f8abda11c2f82ecbf8b373b372e423e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5X1y1LBas5NiuNtfxpdGJ7HgD2TfEVBcx+3ffDwPjGRz8tTHDxtpK/ja9xRV7yDPnrZeYZ8YnUZU2ovmJ21yCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T22:34:09.358391Z","bundle_sha256":"8fb12bb49f0710361a0661e4f70234e112be045dcb7254db54e2a50e4aedb961"}}