{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:VZO5F6QNM37IWNEH6LDWN7AYQV","short_pith_number":"pith:VZO5F6QN","canonical_record":{"source":{"id":"2501.06709","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-01-12T04:29:39Z","cross_cats_sorted":[],"title_canon_sha256":"72bcdf1356ca29a6734a368c6ee7e32c5ee44b4b6b14dd7fb15f02ed3e5c9c5d","abstract_canon_sha256":"91bbb2ee54d0e6c261d5057da5efa91bcb5ca079bff29c62de05265b7d6f727a"},"schema_version":"1.0"},"canonical_sha256":"ae5dd2fa0d66fe8b3487f2c766fc188570104164384c0babc47d81003ee5d702","source":{"kind":"arxiv","id":"2501.06709","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.06709","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"arxiv_version","alias_value":"2501.06709v1","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.06709","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"pith_short_12","alias_value":"VZO5F6QNM37I","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"pith_short_16","alias_value":"VZO5F6QNM37IWNEH","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"pith_short_8","alias_value":"VZO5F6QN","created_at":"2026-07-05T10:00:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:VZO5F6QNM37IWNEH6LDWN7AYQV","target":"record","payload":{"canonical_record":{"source":{"id":"2501.06709","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-01-12T04:29:39Z","cross_cats_sorted":[],"title_canon_sha256":"72bcdf1356ca29a6734a368c6ee7e32c5ee44b4b6b14dd7fb15f02ed3e5c9c5d","abstract_canon_sha256":"91bbb2ee54d0e6c261d5057da5efa91bcb5ca079bff29c62de05265b7d6f727a"},"schema_version":"1.0"},"canonical_sha256":"ae5dd2fa0d66fe8b3487f2c766fc188570104164384c0babc47d81003ee5d702","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:00:08.929388Z","signature_b64":"lsD0Na9aq8IUcLEEC5GGthMq/p8FNjO63nf2y0ibs69WAP8B4rgIP4TQVtVtsLsUa7pJS223b/qNbT7JHDVICg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ae5dd2fa0d66fe8b3487f2c766fc188570104164384c0babc47d81003ee5d702","last_reissued_at":"2026-07-05T10:00:08.928728Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:00:08.928728Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.06709","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":"2026-07-05T10:00:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2MPwxmo2/yzN6eM7crb7g9VWfD9zNYo+/UihLYPFhGzhdPoYFqXicbdiiRMj4bj/YGQHWdT9zRTCDG/32cXIBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T09:39:29.431022Z"},"content_sha256":"e0fb4e984a07bd11a8ed9d063bf3fc3b6340aea427cad92b3ed3a20b3d101605","schema_version":"1.0","event_id":"sha256:e0fb4e984a07bd11a8ed9d063bf3fc3b6340aea427cad92b3ed3a20b3d101605"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:VZO5F6QNM37IWNEH6LDWN7AYQV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Mell: Memory-Efficient Large Language Model Serving via Multi-GPU KV Cache Management","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Chen Fahao, Guo Song, Hong Zicong, Li Peng, Liu Qianli","submitted_at":"2025-01-12T04:29:39Z","abstract_excerpt":"Serving large language models (LLMs) for massive users is challenged by the significant memory footprint of the transient state, known as the key-value (KV) cache, which scales with sequence length and number of requests. Instead of renting or buying more expensive GPUs, the load imbalance of the KV cache across GPUs, coupled with recent advances in inter-GPU communication, provides an opportunity to serve more requests via request migration. However, high migration overhead and unpredictable request patterns make it challenging. Therefore, this paper proposes MELL, a memory-efficient LLM serv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.06709","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/2501.06709/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":"2026-07-05T10:00:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SlYOLN/bcizJ6YZUZXRdgIxrv/moHwkl7uOngTmT7mHaaQKFlccQ1Vf441l3Wll2oiLLoazOw4QbjRTXsBP0BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T09:39:29.431811Z"},"content_sha256":"faf784a4e3efdbba28c4859f0878d694763a7640ad9e14553023fc8d3a3f87cd","schema_version":"1.0","event_id":"sha256:faf784a4e3efdbba28c4859f0878d694763a7640ad9e14553023fc8d3a3f87cd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VZO5F6QNM37IWNEH6LDWN7AYQV/bundle.json","state_url":"https://pith.science/pith/VZO5F6QNM37IWNEH6LDWN7AYQV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VZO5F6QNM37IWNEH6LDWN7AYQV/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":"2026-08-16T09:39:29Z","links":{"resolver":"https://pith.science/pith/VZO5F6QNM37IWNEH6LDWN7AYQV","bundle":"https://pith.science/pith/VZO5F6QNM37IWNEH6LDWN7AYQV/bundle.json","state":"https://pith.science/pith/VZO5F6QNM37IWNEH6LDWN7AYQV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VZO5F6QNM37IWNEH6LDWN7AYQV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:VZO5F6QNM37IWNEH6LDWN7AYQV","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":"91bbb2ee54d0e6c261d5057da5efa91bcb5ca079bff29c62de05265b7d6f727a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-01-12T04:29:39Z","title_canon_sha256":"72bcdf1356ca29a6734a368c6ee7e32c5ee44b4b6b14dd7fb15f02ed3e5c9c5d"},"schema_version":"1.0","source":{"id":"2501.06709","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.06709","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"arxiv_version","alias_value":"2501.06709v1","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.06709","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"pith_short_12","alias_value":"VZO5F6QNM37I","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"pith_short_16","alias_value":"VZO5F6QNM37IWNEH","created_at":"2026-07-05T10:00:08Z"},{"alias_kind":"pith_short_8","alias_value":"VZO5F6QN","created_at":"2026-07-05T10:00:08Z"}],"graph_snapshots":[{"event_id":"sha256:faf784a4e3efdbba28c4859f0878d694763a7640ad9e14553023fc8d3a3f87cd","target":"graph","created_at":"2026-07-05T10:00:08Z","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/2501.06709/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Serving large language models (LLMs) for massive users is challenged by the significant memory footprint of the transient state, known as the key-value (KV) cache, which scales with sequence length and number of requests. Instead of renting or buying more expensive GPUs, the load imbalance of the KV cache across GPUs, coupled with recent advances in inter-GPU communication, provides an opportunity to serve more requests via request migration. However, high migration overhead and unpredictable request patterns make it challenging. Therefore, this paper proposes MELL, a memory-efficient LLM serv","authors_text":"Chen Fahao, Guo Song, Hong Zicong, Li Peng, Liu Qianli","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-01-12T04:29:39Z","title":"Mell: Memory-Efficient Large Language Model Serving via Multi-GPU KV Cache Management"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.06709","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:e0fb4e984a07bd11a8ed9d063bf3fc3b6340aea427cad92b3ed3a20b3d101605","target":"record","created_at":"2026-07-05T10:00:08Z","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":"91bbb2ee54d0e6c261d5057da5efa91bcb5ca079bff29c62de05265b7d6f727a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-01-12T04:29:39Z","title_canon_sha256":"72bcdf1356ca29a6734a368c6ee7e32c5ee44b4b6b14dd7fb15f02ed3e5c9c5d"},"schema_version":"1.0","source":{"id":"2501.06709","kind":"arxiv","version":1}},"canonical_sha256":"ae5dd2fa0d66fe8b3487f2c766fc188570104164384c0babc47d81003ee5d702","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ae5dd2fa0d66fe8b3487f2c766fc188570104164384c0babc47d81003ee5d702","first_computed_at":"2026-07-05T10:00:08.928728Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:00:08.928728Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lsD0Na9aq8IUcLEEC5GGthMq/p8FNjO63nf2y0ibs69WAP8B4rgIP4TQVtVtsLsUa7pJS223b/qNbT7JHDVICg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:00:08.929388Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.06709","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e0fb4e984a07bd11a8ed9d063bf3fc3b6340aea427cad92b3ed3a20b3d101605","sha256:faf784a4e3efdbba28c4859f0878d694763a7640ad9e14553023fc8d3a3f87cd"],"state_sha256":"409641f6bea76886185da3f0c518c5b7088facc3d7cd81c9764526224d2f41c4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RWgIoYqj94BJfCUW7aI/eiqWQ6azHNU9+vLa0HsiVux7i/ABJlKqr60+lskHHSVj8mqWFcvA9ri9q1gdK/ChCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T09:39:29.436777Z","bundle_sha256":"6a8c46e18000f6867d880e9da4d975e5631ccec84544e3796bac38b442c70294"}}