{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:7UW7I2EQYNTQVUKV4ATNJQL4AR","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":"fca5b1b64f0985c557c40d3e83ce03d57b000b6e54725b0e75d9d22d0df10392","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-08-15T13:32:25Z","title_canon_sha256":"664f362b8ff30685b11446981a35520b5f5b76c8157678284d329ed28277bcc6"},"schema_version":"1.0","source":{"id":"2408.08147","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.08147","created_at":"2026-07-05T08:55:45Z"},{"alias_kind":"arxiv_version","alias_value":"2408.08147v1","created_at":"2026-07-05T08:55:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.08147","created_at":"2026-07-05T08:55:45Z"},{"alias_kind":"pith_short_12","alias_value":"7UW7I2EQYNTQ","created_at":"2026-07-05T08:55:45Z"},{"alias_kind":"pith_short_16","alias_value":"7UW7I2EQYNTQVUKV","created_at":"2026-07-05T08:55:45Z"},{"alias_kind":"pith_short_8","alias_value":"7UW7I2EQ","created_at":"2026-07-05T08:55:45Z"}],"graph_snapshots":[{"event_id":"sha256:a082e3dbc9db9f2bc561e657be256ba5fbe932330e59455d4bed553e0b900436","target":"graph","created_at":"2026-07-05T08:55:45Z","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/2408.08147/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Serving disaggregated large language models (LLMs) over tens of thousands of xPU devices (GPUs or NPUs) with reliable performance faces multiple challenges. 1) Ignoring the diversity (various prefixes and tidal requests), treating all the prompts in a mixed pool is inadequate. To facilitate the similarity per scenario and minimize the inner mismatch on P/D (prefill and decoding) processing, fine-grained organization is required, dynamically adjusting P/D ratios for better performance. 2) Due to inaccurate estimation on workload (queue status or maintained connections), the global scheduler eas","authors_text":"Cailong Li, Haoliang Cheng, Hefei Guo, Huimin Lin, Jiandong Ding, Jianxun Song, Jiawei Zhu, Li Ma, Mingyang Song, Peiyang Li, Ruizhi Huan, Tao Wang, Tiandeng Wu, Tiantian Dong, Wen Liu, Wenting Zhou, Xiangkun Xu, Xiaojing Li, Xiao You, Xijian Ju, Xing Chu, Yajing Sun, Yibo Jin, Yicheng Shan, Yipeng Ma, Yongsheng Zhang, Yunpeng Ye, Zhengfan Yuan, Zhengyong Zhang, Zhe Wang","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-08-15T13:32:25Z","title":"P/D-Serve: Serving Disaggregated Large Language Model at Scale"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.08147","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":"sha256:366f8d8d3c3024118acefd56b68b7f2c8bab5ed4c5c1aa2650cc64dd27a295b2","target":"record","created_at":"2026-07-05T08:55:45Z","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":"fca5b1b64f0985c557c40d3e83ce03d57b000b6e54725b0e75d9d22d0df10392","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-08-15T13:32:25Z","title_canon_sha256":"664f362b8ff30685b11446981a35520b5f5b76c8157678284d329ed28277bcc6"},"schema_version":"1.0","source":{"id":"2408.08147","kind":"arxiv","version":1}},"canonical_sha256":"fd2df46890c3670ad155e026d4c17c0452e97d4ba1585756454da53f203d3749","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fd2df46890c3670ad155e026d4c17c0452e97d4ba1585756454da53f203d3749","first_computed_at":"2026-07-05T08:55:45.186651Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:55:45.186651Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fWjDkt1a6NjBT0HoGTR/QK1+I2FhJH4PyNSZFbGC1PGkByu7XzJy9pPK3oelvCE34UW0qWtPi+UtuqZO8IWbAg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:55:45.187111Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.08147","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:366f8d8d3c3024118acefd56b68b7f2c8bab5ed4c5c1aa2650cc64dd27a295b2","sha256:a082e3dbc9db9f2bc561e657be256ba5fbe932330e59455d4bed553e0b900436"],"state_sha256":"8e24c93ddafc0659c1a85d08c3048b026355962b94a2e6a6225cf4309086ee49"}