{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,19]],"date-time":"2026-09-19T00:46:54Z","timestamp":1789778814487,"version":"4.0.1"},"publisher-location":"New York, NY, USA","reference-count":42,"publisher":"ACM","license":[{"start":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T00:00:00Z","timestamp":1786406400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"National Natural Science Foundation of China","award":["U25B2032"],"award-info":[{"award-number":["U25B2032"]}]},{"name":"National Natural Science Foundation of China","award":["62572128"],"award-info":[{"award-number":["62572128"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,8,17]]},"DOI":"10.1145\/3789240.3829161","type":"proceedings-article","created":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T18:33:22Z","timestamp":1786473202000},"page":"519-534","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["DistDPU: A Disaggregated DPU Architecture for High-Performance and Cost-Efficient AI Clouds"],"prefix":"10.1145","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-7265-9512","authenticated-orcid":false,"given":"Hao","family":"Mei","sequence":"first","affiliation":[{"name":"Fudan University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-3393-1962","authenticated-orcid":false,"given":"Lizhou","family":"Gao","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-9920-7155","authenticated-orcid":false,"given":"Yuanyi","family":"Zhu","sequence":"additional","affiliation":[{"name":"Fudan University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-2587-5440","authenticated-orcid":false,"given":"Liang","family":"Wang","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4400-2289","authenticated-orcid":false,"given":"Peng","family":"Yang","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-7662-8855","authenticated-orcid":false,"given":"Chao","family":"Pei","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-6286-8964","authenticated-orcid":false,"given":"Chuhao","family":"Chen","sequence":"additional","affiliation":[{"name":"Fudan University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-5821-0025","authenticated-orcid":false,"given":"Zijian","family":"Li","sequence":"additional","affiliation":[{"name":"Fudan University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-3351-786X","authenticated-orcid":false,"given":"Jian","family":"Zhao","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-5235-8722","authenticated-orcid":false,"given":"Dongbo","family":"Gu","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-7819-119X","authenticated-orcid":false,"given":"Hongchen","family":"Ren","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-4678-6221","authenticated-orcid":false,"given":"Jiyuan","family":"Chen","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-2496-5446","authenticated-orcid":false,"given":"Junpeng","family":"Zhang","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-2929-8179","authenticated-orcid":false,"given":"Yunpeng","family":"Guan","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-7676-0967","authenticated-orcid":false,"given":"Jianye","family":"Yuan","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8932-4022","authenticated-orcid":false,"given":"Jian","family":"Wang","sequence":"additional","affiliation":[{"name":"Tencent, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9215-4298","authenticated-orcid":false,"given":"Yibo","family":"Huang","sequence":"additional","affiliation":[{"name":"University of Michigan, Ann Arbor, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0958-8547","authenticated-orcid":false,"given":"Yang","family":"Xu","sequence":"additional","affiliation":[{"name":"Fudan University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,8,11]]},"reference":[{"key":"e_1_3_2_1_1_1","unstructured":"Advanced Micro Devices Inc. 2023. 4th Gen AMD EPYC Processor Architecture (third edition ed.). AMD Santa Clara CA. https:\/\/www.amd.com\/content\/dam\/amd\/en\/documents\/products\/epyc\/4th-gen-epyc-processor-architecture-white-paper.pdf."},{"key":"e_1_3_2_1_2_1","volume-title":"The Security Design of the AWS Nitro System. Amazon Web Services","author":"Services Amazon Web","unstructured":"Amazon Web Services. 2024. The Security Design of the AWS Nitro System. Amazon Web Services, Seattle, WA. AWS Whitepaper. https:\/\/docs.aws.amazon.com\/whitepapers\/latest\/security-design-of-aws-nitro-system\/security-design-of-aws-nitro-system.html."},{"key":"e_1_3_2_1_3_1","first-page":"25","article-title":"Warehouse-scale computing: Entering the teenage decade. ISCA Keynote","volume":"12","author":"Barroso Luiz Andre","year":"2011","unstructured":"Luiz Andre Barroso. 2011. Warehouse-scale computing: Entering the teenage decade. ISCA Keynote, SJ 12 (2011), 25\u201360.","journal-title":"SJ"},{"key":"e_1_3_2_1_4_1","volume-title":"11th USENIX Symposium on Operating Systems Design and Implementation (OSDI 14)","author":"Belay Adam","year":"2014","unstructured":"Adam Belay, George Prekas, Ana Klimovic, Samuel Grossman, Christos Kozyrakis, and Edouard Bugnion. 2014. IX: a protected dataplane operating system for high throughput and low latency. In 11th USENIX Symposium on Operating Systems Design and Implementation (OSDI 14). 49\u201365."},{"key":"e_1_3_2_1_5_1","volume-title":"NVIDIA Data Center Processing Unit (DPU) Architecture. In 2021 IEEE Hot Chips 33 Symposium (HCS). IEEE, 1\u201320","author":"Burstein Idan","year":"2021","unstructured":"Idan Burstein. 2021. NVIDIA Data Center Processing Unit (DPU) Architecture. In 2021 IEEE Hot Chips 33 Symposium (HCS). IEEE, 1\u201320."},{"key":"e_1_3_2_1_6_1","unstructured":"ByteDance. 2025. ByteDance veRoCE Transport Protocol. Technical Report. ByteDance. https:\/\/developer.volcengine.com\/resource\/7584346532149723178"},{"key":"e_1_3_2_1_7_1","volume-title":"15th USENIX symposium on networked systems design and implementation (NSDI 18)","author":"Dalton Michael","year":"2018","unstructured":"Michael Dalton, David Schultz, Jacob Adriaens, Ahsan Arefin, Anshuman Gupta, Brian Fahs, Dima Rubinstein, Enrique Cauich Zermeno, Erik Rubow, James Alexander Docauer, et al. 2018. Andromeda: Performance, isolation, and velocity at scale in cloud network virtualization. In 15th USENIX symposium on networked systems design and implementation (NSDI 18). 373\u2013387."},{"key":"e_1_3_2_1_8_1","unstructured":"Abhimanyu Dubey Abhinav Jauhri Abhinav Pandey Abhishek Kadian Ahmad Al-Dahle Aiesha Letman Akhil Mathur Alan Schelten Amy Yang Angela Fan et al. 2024. The Llama 3 Herd of Models. arXiv e-prints (2024) arXiv-2407."},{"key":"e_1_3_2_1_9_1","volume-title":"15th USENIX Symposium on Networked Systems Design and Implementation (NSDI 18)","author":"Firestone Daniel","year":"2018","unstructured":"Daniel Firestone, Andrew Putnam, Sambhrama Mundkur, Derek Chiou, Alireza Dabagh, Mike Andrewartha, Hari Angepat, Vivek Bhanu, Adrian Caulfield, Eric Chung, et al. 2018. Azure accelerated networking:SmartNICs in the public cloud. In 15th USENIX Symposium on Networked Systems Design and Implementation (NSDI 18). 51\u201366."},{"key":"e_1_3_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1145\/3651890.3672233"},{"key":"e_1_3_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1145\/3503222.3507762"},{"key":"e_1_3_2_1_12_1","volume-title":"Supplement to InfiniBand Architecture Specification","author":"InfiniBand Trade Association (IBTA) 2014.","unstructured":"InfiniBand Trade Association (IBTA) 2014. Supplement to InfiniBand Architecture Specification Volume 1 Release 1.2.1, Annex A17: RoCEv2. InfiniBand Trade Association (IBTA). https:\/\/cw.infinibandta.org\/document\/dl\/7781"},{"key":"e_1_3_2_1_13_1","unstructured":"Intel Corporation. 2020. Intel\u00ae Scalable I\/O Virtualization Technical Specification. Intel Corporation. https:\/\/www.intel.com\/content\/www\/us\/en\/content-details\/671403\/intel-scalable-i-o-virtualization-technical-specification.html Order Number: 337679-002 Revision 1.1."},{"key":"e_1_3_2_1_14_1","volume-title":"Analysis of Large-Scale Multi-Tenant GPU Clusters for DNN Training Workloads. In 2019 USENIX Annual Technical Conference (USENIX ATC 19)","author":"Jeon Myeongjae","year":"2019","unstructured":"Myeongjae Jeon, Shivaram Venkataraman, Amar Phanishayee, Junjie Qian, Wencong Xiao, and Fan Yang. 2019. Analysis of Large-Scale Multi-Tenant GPU Clusters for DNN Training Workloads. In 2019 USENIX Annual Technical Conference (USENIX ATC 19). USENIX Association, 947\u2013960."},{"key":"e_1_3_2_1_15_1","volume-title":"11th USENIX symposium on networked systems design and implementation (NSDI 14)","author":"Koponen Teemu","year":"2014","unstructured":"Teemu Koponen, Keith Amidon, Peter Balland, Mart\u00edn Casado, Anupam Chanda, Bryan Fulton, Igor Ganichev, Jesse Gross, Paul Ingram, Ethan Jackson, et al. 2014. Network virtualization in multi-tenant datacenters. In 11th USENIX symposium on networked systems design and implementation (NSDI 14). 203\u2013216."},{"key":"e_1_3_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1145\/3387514.3406591"},{"key":"e_1_3_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"e_1_3_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/2934872.2934897"},{"key":"e_1_3_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3341302.3342085"},{"key":"e_1_3_2_1_20_1","unstructured":"Aixin Liu Bei Feng Bing Xue Bingxuan Wang Bochao Wu Chengda Lu Chenggang Zhao Chengqi Deng Chenyu Zhang Chong Ruan et al. 2024. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437 (2024)."},{"key":"e_1_3_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/3718958.3750496"},{"key":"e_1_3_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/3718958.3750539"},{"key":"e_1_3_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359657"},{"key":"e_1_3_2_1_24_1","unstructured":"NVIDIA. 2025. ConnectX-7 400G Adapters. https:\/\/resources.nvidia.com\/en-us-accelerated-networking-resource-library\/connectx-7-datasheet."},{"key":"e_1_3_2_1_25_1","unstructured":"NVIDIA. 2025. NVIDIA BlueField-3 Networking Platform. https:\/\/resources.nvidia.com\/en-us-accelerated-networking-resource-library\/datasheet-nvidia-bluefield."},{"key":"e_1_3_2_1_26_1","unstructured":"NVIDIA Corporation. 2024. TensorRT-LLM GitHub Repository. https:\/\/github.com\/NVIDIA\/TensorRT-LLM. Accessed: 2026-02-07."},{"key":"e_1_3_2_1_27_1","volume-title":"19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22)","author":"Olteanu Vladimir","year":"2022","unstructured":"Vladimir Olteanu, Haggai Eran, Dragos Dumitrescu, Adrian Popa, Cristi Baciu, Mark Silberstein, Georgios Nikolaidis, Mark Handley, and Costin Raiciu. 2022. An edge-queued datagram service for all datacenter traffic. In 19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22). 761\u2013777."},{"key":"e_1_3_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA59077.2024.00019"},{"key":"e_1_3_2_1_29_1","volume-title":"13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18)","author":"Phothilimthana Phitchaya Mangpo","year":"2018","unstructured":"Phitchaya Mangpo Phothilimthana, Ming Liu, Antoine Kaufmann, Simon Peter, Rastislav Bodik, and Thomas Anderson. 2018. Floem: A programming system for NIC-Accelerated network applications. In 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18). 663\u2013679."},{"key":"e_1_3_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1145\/2678373.2665678"},{"key":"e_1_3_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/ISPASS48437.2020.00018"},{"key":"e_1_3_2_1_32_1","volume-title":"Megatron-LM: Training multi-billion parameter language models using model parallelism. arXiv preprint arXiv:1909.08053","author":"Shoeybi Mohammad","year":"2019","unstructured":"Mohammad Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper, and Bryan Catanzaro. 2019. Megatron-LM: Training multi-billion parameter language models using model parallelism. arXiv preprint arXiv:1909.08053 (2019)."},{"key":"e_1_3_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1145\/3387514.3405897"},{"key":"e_1_3_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1145\/3718958.3754353"},{"key":"e_1_3_2_1_35_1","volume-title":"18th USENIX Symposium on Networked Systems Design and Implementation (NSDI 21)","author":"Sultana Nik","year":"2021","unstructured":"Nik Sultana, John Sonchack, Hans Giesen, Isaac Pedisich, Zhaoyang Han, Nishanth Shyamkumar, Shivani Burad, Andr\u00e9 DeHon, and Boon Thau Loo. 2021. Flightplan: Dataplane disaggregation and placement for p4 programs. In 18th USENIX Symposium on Networked Systems Design and Implementation (NSDI 21). 571\u2013592."},{"key":"e_1_3_2_1_36_1","volume-title":"14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20)","author":"Tang Chunqiang","year":"2020","unstructured":"Chunqiang Tang, Kenny Yu, Kaushik Veeraraghavan, Jonathan Kaldor, Scott Michelson, Thawan Kooburat, Aravind Anbudurai, Matthew Clark, Kabir Gogia, Long Cheng, et al. 2020. Twine: A unified cluster management system for shared infrastructure. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20). 787\u2013803."},{"key":"e_1_3_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/3718958.3750501"},{"key":"e_1_3_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1145\/3544216.3544271"},{"key":"e_1_3_2_1_39_1","volume-title":"16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22)","author":"Zheng Lianmin","year":"2022","unstructured":"Lianmin Zheng, Zhuohan Li, Hao Zhang, Yonghao Zhuang, Zhifeng Chen, Yanping Huang, Yida Wang, Yuanzhong Xu, Danyang Zhuo, Eric P Xing, et al. 2022. Alpa: Automating inter-and Intra-Operator parallelism for distributed deep learning. In 16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22). 559\u2013578."},{"key":"e_1_3_2_1_40_1","volume-title":"Jeff Huang, Cody Hao Yu, Shiyi Cao, Christos Kozyrakis, Ion Stoica, Joseph E Gonzalez, et al.","author":"Zheng Lianmin","year":"2024","unstructured":"Lianmin Zheng, Liangsheng Yin, Zhiqiang Xie, Chuyue Livia Sun, Jeff Huang, Cody Hao Yu, Shiyi Cao, Christos Kozyrakis, Ion Stoica, Joseph E Gonzalez, et al. 2024. Sglang: Efficient execution of structured language model programs. Advances in neural information processing systems 37 (2024), 62557\u201362583."},{"key":"e_1_3_2_1_41_1","volume-title":"18th USENIX Symposium on Operating Systems Design and Implementation (OSDI 24)","author":"Zhong Yinmin","year":"2024","unstructured":"Yinmin Zhong, Shengyu Liu, Junda Chen, Jianbo Hu, Yibo Zhu, Xuanzhe Liu, Xin Jin, and Hao Zhang. 2024. DistServe: Disaggregating prefill and decoding for goodput-optimized large language model serving. In 18th USENIX Symposium on Operating Systems Design and Implementation (OSDI 24). 193\u2013210."},{"key":"e_1_3_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.1145\/2829988.2787484"}],"event":{"name":"SIGCOMM '26: ACM SIGCOMM 2026 Conference","location":"Colorado Convention Center Denver CO USA","acronym":"SIGCOMM '26","sponsor":["SIGCOMM ACM Special Interest Group on Data Communication"]},"container-title":["Proceedings of the ACM SIGCOMM 2026 Conference"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3789240.3829161","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T18:37:44Z","timestamp":1786473464000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3789240.3829161"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,11]]},"references-count":42,"alternative-id":["10.1145\/3789240.3829161","10.1145\/3789240"],"URL":"https:\/\/doi.org\/10.1145\/3789240.3829161","relation":{},"subject":[],"published":{"date-parts":[[2026,8,11]]},"assertion":[{"value":"2026-08-11","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}