{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:Z7TPZI636O5ZOQCTSY7RB6FXP7","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":"f4ec8e77cd5d9922d1d63fadfc6c777b9572c2d21d5ea53ea71f12cd889bf479","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-08-22T23:00:40Z","title_canon_sha256":"f0dd478d1c47e335eea933aded09020cb48ea72d19c3611eeb1cd10739ceb9df"},"schema_version":"1.0","source":{"id":"2408.12757","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.12757","created_at":"2026-07-05T11:08:59Z"},{"alias_kind":"arxiv_version","alias_value":"2408.12757v2","created_at":"2026-07-05T11:08:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.12757","created_at":"2026-07-05T11:08:59Z"},{"alias_kind":"pith_short_12","alias_value":"Z7TPZI636O5Z","created_at":"2026-07-05T11:08:59Z"},{"alias_kind":"pith_short_16","alias_value":"Z7TPZI636O5ZOQCT","created_at":"2026-07-05T11:08:59Z"},{"alias_kind":"pith_short_8","alias_value":"Z7TPZI63","created_at":"2026-07-05T11:08:59Z"}],"graph_snapshots":[{"event_id":"sha256:3dee6baef7ed450094a993aa2153c5d0cf60dcfe224569fdd0d235de7203d783","target":"graph","created_at":"2026-07-05T11:08:59Z","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/2408.12757/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have resulted in a surging demand for planet-scale serving systems, where tens of thousands of GPUs continuously serve hundreds of millions of users. Consequently, throughput has emerged as a key metric that determines serving systems' performance. Due to large model sizes and memory-intensive self-attention, LLM serving has been commonly assumed to be memory-bound. Through a detailed analysis, we show that despite having memory-intensive components, end-to-end LLM serving is compute bound for most common workloads and LLMs. Alas, most existing serving engines fall","authors_text":"Arvind Krishnamurthy, Baris Kasikci, Chien-Yu Lin, Dedong Xie, Gefei Zuo, Kan Zhu, Keisuke Kamahori, Liangyu Zhao, Qinyu Xu, Stephanie Wang, Tian Tang, Yile Gu, Yilong Zhao, Yufei Gao, Zihao Ye, Ziren Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-08-22T23:00:40Z","title":"NanoFlow: Towards Optimal Large Language Model Serving Throughput"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.12757","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":"sha256:9f1e79824b1fab30bee43f54440f63bfb997acc7fdd8a887bcdfb147bef74b36","target":"record","created_at":"2026-07-05T11:08:59Z","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":"f4ec8e77cd5d9922d1d63fadfc6c777b9572c2d21d5ea53ea71f12cd889bf479","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-08-22T23:00:40Z","title_canon_sha256":"f0dd478d1c47e335eea933aded09020cb48ea72d19c3611eeb1cd10739ceb9df"},"schema_version":"1.0","source":{"id":"2408.12757","kind":"arxiv","version":2}},"canonical_sha256":"cfe6fca3dbf3bb974053963f10f8b77fd89d7633fed379f0e77d497353f568ec","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cfe6fca3dbf3bb974053963f10f8b77fd89d7633fed379f0e77d497353f568ec","first_computed_at":"2026-07-05T11:08:59.562602Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:08:59.562602Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2SD0VonSjXuiJeHmQ9KA5w5/0mJJWDf922AT10DbmRCpDdot9fPXoVM+XURNF/Y+E0mTi1EG/5Iw+xuXIZdWDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:08:59.563122Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.12757","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9f1e79824b1fab30bee43f54440f63bfb997acc7fdd8a887bcdfb147bef74b36","sha256:3dee6baef7ed450094a993aa2153c5d0cf60dcfe224569fdd0d235de7203d783"],"state_sha256":"c4c50ae5899620d1eceb8fd205c3fe18ccc2d2933a22cebcdab0c9190037024f"}