{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:Y4EOI4VS4I4B7DDUME5IGWMWH2","short_pith_number":"pith:Y4EOI4VS","canonical_record":{"source":{"id":"2403.01876","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-04T09:32:05Z","cross_cats_sorted":[],"title_canon_sha256":"a18ea48afb22b447fb0e0f0aa2e558424ee97ddd4eae33c4edf0da655b10000f","abstract_canon_sha256":"3e0ae57aaf335501062feefc9cc9a17ae256e08d06f1ee360bcf61a9cdbd6f77"},"schema_version":"1.0"},"canonical_sha256":"c708e472b2e2381f8c74613a8359963e88561ca32dabbebca9f4fa6e47055d49","source":{"kind":"arxiv","id":"2403.01876","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.01876","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"arxiv_version","alias_value":"2403.01876v1","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.01876","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"pith_short_12","alias_value":"Y4EOI4VS4I4B","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"pith_short_16","alias_value":"Y4EOI4VS4I4B7DDU","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"pith_short_8","alias_value":"Y4EOI4VS","created_at":"2026-07-05T07:51:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:Y4EOI4VS4I4B7DDUME5IGWMWH2","target":"record","payload":{"canonical_record":{"source":{"id":"2403.01876","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-04T09:32:05Z","cross_cats_sorted":[],"title_canon_sha256":"a18ea48afb22b447fb0e0f0aa2e558424ee97ddd4eae33c4edf0da655b10000f","abstract_canon_sha256":"3e0ae57aaf335501062feefc9cc9a17ae256e08d06f1ee360bcf61a9cdbd6f77"},"schema_version":"1.0"},"canonical_sha256":"c708e472b2e2381f8c74613a8359963e88561ca32dabbebca9f4fa6e47055d49","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:51:53.386892Z","signature_b64":"PRUPAXgYxuLHyVFTWXV8fNfUfWeoq4aCks5Tgr2B9EKlhhT/ytRvooYl2wk+HykxjvNPDpc5pU/xjIpA56BoDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c708e472b2e2381f8c74613a8359963e88561ca32dabbebca9f4fa6e47055d49","last_reissued_at":"2026-07-05T07:51:53.386375Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:51:53.386375Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.01876","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":"2026-07-05T07:51:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T4/tvaEBexWp73iVGllK9comEG30A+4IRpKjECCFB3PQ7eM1InkIxX6WBK2vUSVHtdQh9712nSQtV/yQKJ2IDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T11:12:00.539350Z"},"content_sha256":"fc370add5d31e654212c746bc80aecb1a89953b4b81777973e353b2ce6e97848","schema_version":"1.0","event_id":"sha256:fc370add5d31e654212c746bc80aecb1a89953b4b81777973e353b2ce6e97848"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:Y4EOI4VS4I4B7DDUME5IGWMWH2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"D\\'ej\\`aVu: KV-cache Streaming for Fast, Fault-tolerant Generative LLM Serving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Amar Phanishayee, Ana Klimovic, Foteini Strati, Jakub Tarnawski, Sara McAllister","submitted_at":"2024-03-04T09:32:05Z","abstract_excerpt":"Distributed LLM serving is costly and often underutilizes hardware accelerators due to three key challenges: bubbles in pipeline-parallel deployments caused by the bimodal latency of prompt and token processing, GPU memory overprovisioning, and long recovery times in case of failures. In this paper, we propose D\\'ej\\`aVu, a system to address all these challenges using a versatile and efficient KV cache streaming library (D\\'ej\\`aVuLib). Using D\\'ej\\`aVuLib, we propose and implement efficient prompt-token disaggregation to reduce pipeline bubbles, microbatch swapping for efficient GPU memory ma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.01876","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/2403.01876/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":"2026-07-05T07:51:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7w2NayQDqxRAyxyikc6Ps61INoSKZb1+/K9jFmH1foaZxyY2+NGmQu4xkf7ywYLcm/B+wh1uCV/v8KBt3MfmBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T11:12:00.540334Z"},"content_sha256":"b62920e75ca0f9950d6768995f5ea9829e1edbd7e22dc2adebd52494eab86271","schema_version":"1.0","event_id":"sha256:b62920e75ca0f9950d6768995f5ea9829e1edbd7e22dc2adebd52494eab86271"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Y4EOI4VS4I4B7DDUME5IGWMWH2/bundle.json","state_url":"https://pith.science/pith/Y4EOI4VS4I4B7DDUME5IGWMWH2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Y4EOI4VS4I4B7DDUME5IGWMWH2/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":"2026-08-09T11:12:00Z","links":{"resolver":"https://pith.science/pith/Y4EOI4VS4I4B7DDUME5IGWMWH2","bundle":"https://pith.science/pith/Y4EOI4VS4I4B7DDUME5IGWMWH2/bundle.json","state":"https://pith.science/pith/Y4EOI4VS4I4B7DDUME5IGWMWH2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Y4EOI4VS4I4B7DDUME5IGWMWH2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:Y4EOI4VS4I4B7DDUME5IGWMWH2","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":"3e0ae57aaf335501062feefc9cc9a17ae256e08d06f1ee360bcf61a9cdbd6f77","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-04T09:32:05Z","title_canon_sha256":"a18ea48afb22b447fb0e0f0aa2e558424ee97ddd4eae33c4edf0da655b10000f"},"schema_version":"1.0","source":{"id":"2403.01876","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.01876","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"arxiv_version","alias_value":"2403.01876v1","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.01876","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"pith_short_12","alias_value":"Y4EOI4VS4I4B","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"pith_short_16","alias_value":"Y4EOI4VS4I4B7DDU","created_at":"2026-07-05T07:51:53Z"},{"alias_kind":"pith_short_8","alias_value":"Y4EOI4VS","created_at":"2026-07-05T07:51:53Z"}],"graph_snapshots":[{"event_id":"sha256:b62920e75ca0f9950d6768995f5ea9829e1edbd7e22dc2adebd52494eab86271","target":"graph","created_at":"2026-07-05T07:51:53Z","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/2403.01876/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Distributed LLM serving is costly and often underutilizes hardware accelerators due to three key challenges: bubbles in pipeline-parallel deployments caused by the bimodal latency of prompt and token processing, GPU memory overprovisioning, and long recovery times in case of failures. In this paper, we propose D\\'ej\\`aVu, a system to address all these challenges using a versatile and efficient KV cache streaming library (D\\'ej\\`aVuLib). Using D\\'ej\\`aVuLib, we propose and implement efficient prompt-token disaggregation to reduce pipeline bubbles, microbatch swapping for efficient GPU memory ma","authors_text":"Amar Phanishayee, Ana Klimovic, Foteini Strati, Jakub Tarnawski, Sara McAllister","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-04T09:32:05Z","title":"D\\'ej\\`aVu: KV-cache Streaming for Fast, Fault-tolerant Generative LLM Serving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.01876","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:fc370add5d31e654212c746bc80aecb1a89953b4b81777973e353b2ce6e97848","target":"record","created_at":"2026-07-05T07:51:53Z","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":"3e0ae57aaf335501062feefc9cc9a17ae256e08d06f1ee360bcf61a9cdbd6f77","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-03-04T09:32:05Z","title_canon_sha256":"a18ea48afb22b447fb0e0f0aa2e558424ee97ddd4eae33c4edf0da655b10000f"},"schema_version":"1.0","source":{"id":"2403.01876","kind":"arxiv","version":1}},"canonical_sha256":"c708e472b2e2381f8c74613a8359963e88561ca32dabbebca9f4fa6e47055d49","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c708e472b2e2381f8c74613a8359963e88561ca32dabbebca9f4fa6e47055d49","first_computed_at":"2026-07-05T07:51:53.386375Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:51:53.386375Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PRUPAXgYxuLHyVFTWXV8fNfUfWeoq4aCks5Tgr2B9EKlhhT/ytRvooYl2wk+HykxjvNPDpc5pU/xjIpA56BoDw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:51:53.386892Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.01876","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fc370add5d31e654212c746bc80aecb1a89953b4b81777973e353b2ce6e97848","sha256:b62920e75ca0f9950d6768995f5ea9829e1edbd7e22dc2adebd52494eab86271"],"state_sha256":"0a5842cd8a2dda1985a8247fb544186e66807fabc75f4afb890b2473229602b0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OL1UYsI1I2fU1AiamcVgt/clgteXZTHZ52+Zl9D/ai99EXrI7BiN84hhx0NKd5IVpRF1vAHU9OuUx6zDUk9YAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T11:12:00.547531Z","bundle_sha256":"f87067c0e0019bcddaf27732270537963f3a5e6e42e546836447f0ad8d8baa29"}}