{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XSSOFW5MNP66E2GWAVN3DV5HCB","short_pith_number":"pith:XSSOFW5M","schema_version":"1.0","canonical_sha256":"bca4e2dbac6bfde268d6055bb1d7a71057668fdc904763ae58f025b53b97b1ed","source":{"kind":"arxiv","id":"2408.14158","version":2},"attestation_state":"computed","paper":{"title":"Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Chenggang Zhao, Chengqi Deng, Guanting Chen, Haowei Zhang, Honghui Ding, Hui Qu, Jianzhong Guo, Jiashi Li, Junjie Qiu, Kai Dong, Kang Guan, Liyue Zhang, Mingchuan Zhang, Minghui Tang, Panpan Huang, Qiushi Du, Shanghao Lu, Shangyan Zhou, Shanhuang Chen, Shengfeng Ye, Shiyu Wang, Shuiping Yu, Shunfeng Zhou, Tian Pei, Wei An, Wenfeng Liang, Wenjun Gao, Wentao Zhang, Xiao Bi, XiaoDong Liu, Xiaotao Nie, Xiaowen Sun, Xin Liu, Xuan Lu, Xuecheng Su, Yanhong Xu, Yaohui Wang, Yao Zhao, Yichao Zhang, Yiliang Xiong, Ying He, Yixuan Tan, Yiyuan Liu, Yongji Wang, Yongqiang Guo, Yuheng Zou, Yukun Zha, Yuxuan Liu, Zehui Ren, Zhangli Sha, Zhe Fu, Ziwei Xie","submitted_at":"2024-08-26T10:11:56Z","abstract_excerpt":"The rapid progress in Deep Learning (DL) and Large Language Models (LLMs) has exponentially increased demands of computational power and bandwidth. This, combined with the high costs of faster computing chips and interconnects, has significantly inflated High Performance Computing (HPC) construction costs. To address these challenges, we introduce the Fire-Flyer AI-HPC architecture, a synergistic hardware-software co-design framework and its best practices. For DL training, we deployed the Fire-Flyer 2 with 10,000 PCIe A100 GPUs, achieved performance approximating the DGX-A100 while reducing c"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2408.14158","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-08-26T10:11:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"583e92dd72e4a087d72c7756debf8032fb798bc87d657bb2bfd4cd6667d3eab9","abstract_canon_sha256":"be2d2407fadffa1f5c81dba03bfd6e4c5907d57b3c9ca3317aba4b6c800d7a41"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:01:38.364868Z","signature_b64":"WUJCwfhe0RcNHHDLzs9/9GTLa+4+vtdW6mwOze6jvCkqO1v+fLJ2l03Lml+9uwxMNekXfdxSoQ24dv1ZcSqIAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bca4e2dbac6bfde268d6055bb1d7a71057668fdc904763ae58f025b53b97b1ed","last_reissued_at":"2026-07-05T09:01:38.364337Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:01:38.364337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DC","authors_text":"Chenggang Zhao, Chengqi Deng, Guanting Chen, Haowei Zhang, Honghui Ding, Hui Qu, Jianzhong Guo, Jiashi Li, Junjie Qiu, Kai Dong, Kang Guan, Liyue Zhang, Mingchuan Zhang, Minghui Tang, Panpan Huang, Qiushi Du, Shanghao Lu, Shangyan Zhou, Shanhuang Chen, Shengfeng Ye, Shiyu Wang, Shuiping Yu, Shunfeng Zhou, Tian Pei, Wei An, Wenfeng Liang, Wenjun Gao, Wentao Zhang, Xiao Bi, XiaoDong Liu, Xiaotao Nie, Xiaowen Sun, Xin Liu, Xuan Lu, Xuecheng Su, Yanhong Xu, Yaohui Wang, Yao Zhao, Yichao Zhang, Yiliang Xiong, Ying He, Yixuan Tan, Yiyuan Liu, Yongji Wang, Yongqiang Guo, Yuheng Zou, Yukun Zha, Yuxuan Liu, Zehui Ren, Zhangli Sha, Zhe Fu, Ziwei Xie","submitted_at":"2024-08-26T10:11:56Z","abstract_excerpt":"The rapid progress in Deep Learning (DL) and Large Language Models (LLMs) has exponentially increased demands of computational power and bandwidth. This, combined with the high costs of faster computing chips and interconnects, has significantly inflated High Performance Computing (HPC) construction costs. To address these challenges, we introduce the Fire-Flyer AI-HPC architecture, a synergistic hardware-software co-design framework and its best practices. For DL training, we deployed the Fire-Flyer 2 with 10,000 PCIe A100 GPUs, achieved performance approximating the DGX-A100 while reducing c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.14158","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/2408.14158/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2408.14158","created_at":"2026-07-05T09:01:38.364394+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.14158v2","created_at":"2026-07-05T09:01:38.364394+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.14158","created_at":"2026-07-05T09:01:38.364394+00:00"},{"alias_kind":"pith_short_12","alias_value":"XSSOFW5MNP66","created_at":"2026-07-05T09:01:38.364394+00:00"},{"alias_kind":"pith_short_16","alias_value":"XSSOFW5MNP66E2GW","created_at":"2026-07-05T09:01:38.364394+00:00"},{"alias_kind":"pith_short_8","alias_value":"XSSOFW5M","created_at":"2026-07-05T09:01:38.364394+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.05450","citing_title":"Decentralized Diffusion Models","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XSSOFW5MNP66E2GWAVN3DV5HCB","json":"https://pith.science/pith/XSSOFW5MNP66E2GWAVN3DV5HCB.json","graph_json":"https://pith.science/api/pith-number/XSSOFW5MNP66E2GWAVN3DV5HCB/graph.json","events_json":"https://pith.science/api/pith-number/XSSOFW5MNP66E2GWAVN3DV5HCB/events.json","paper":"https://pith.science/paper/XSSOFW5M"},"agent_actions":{"view_html":"https://pith.science/pith/XSSOFW5MNP66E2GWAVN3DV5HCB","download_json":"https://pith.science/pith/XSSOFW5MNP66E2GWAVN3DV5HCB.json","view_paper":"https://pith.science/paper/XSSOFW5M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.14158&json=true","fetch_graph":"https://pith.science/api/pith-number/XSSOFW5MNP66E2GWAVN3DV5HCB/graph.json","fetch_events":"https://pith.science/api/pith-number/XSSOFW5MNP66E2GWAVN3DV5HCB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XSSOFW5MNP66E2GWAVN3DV5HCB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XSSOFW5MNP66E2GWAVN3DV5HCB/action/storage_attestation","attest_author":"https://pith.science/pith/XSSOFW5MNP66E2GWAVN3DV5HCB/action/author_attestation","sign_citation":"https://pith.science/pith/XSSOFW5MNP66E2GWAVN3DV5HCB/action/citation_signature","submit_replication":"https://pith.science/pith/XSSOFW5MNP66E2GWAVN3DV5HCB/action/replication_record"}},"created_at":"2026-07-05T09:01:38.364394+00:00","updated_at":"2026-07-05T09:01:38.364394+00:00"}