{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T07:14:11Z","timestamp":1772694851396,"version":"3.50.1"},"reference-count":93,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,1,31]],"date-time":"2026-01-31T00:00:00Z","timestamp":1769817600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,31]],"date-time":"2026-01-31T00:00:00Z","timestamp":1769817600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100018537","name":"National Science and Technology Major Project","doi-asserted-by":"publisher","award":["2022ZD0115200"],"award-info":[{"award-number":["2022ZD0115200"]}],"id":[{"id":"10.13039\/501100018537","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"NSFC","doi-asserted-by":"publisher","award":["62502255,62125403,U24A20234,92464302,U24B20164"],"award-info":[{"award-number":["62502255,62125403,U24A20234,92464302,U24B20164"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003399","name":"Shanghai Municipal Science and Technology Major Project","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100003399","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004608","name":"Natural Science Foundation of Jiangsu Province Basic Research Program","doi-asserted-by":"publisher","award":["BK20243042"],"award-info":[{"award-number":["BK20243042"]}],"id":[{"id":"10.13039\/501100004608","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100017582","name":"Beijing National Research Center for Information Science and Technology","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100017582","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100014380","name":"Beijing Advanced Innovation Center for Integrated Circuits","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100014380","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,1,31]]},"DOI":"10.1109\/hpca68181.2026.11408518","type":"proceedings-article","created":{"date-parts":[[2026,3,4]],"date-time":"2026-03-04T20:47:22Z","timestamp":1772657242000},"page":"1-16","source":"Crossref","is-referenced-by-count":0,"title":["FACE: Fully Overlapped PD Scheduling and Multi-Level Architecture Co-Exploration on Wafer"],"prefix":"10.1109","author":[{"given":"Zheng","family":"Xu","sequence":"first","affiliation":[{"name":"School of Integrated Circuits, BNRist, Tsinghua University,Beijing,China,100084"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dehao","family":"Kong","sequence":"additional","affiliation":[{"name":"School of Integrated Circuits, BNRist, Tsinghua University,Beijing,China,100084"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiaxin","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Integrated Circuits, BNRist, Tsinghua University,Beijing,China,100084"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dingcheng","family":"Jiang","sequence":"additional","affiliation":[{"name":"School of Integrated Circuits, BNRist, Tsinghua University,Beijing,China,100084"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xu","family":"Dai","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory,Shanghai,China,200433"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jinyi","family":"Deng","sequence":"additional","affiliation":[{"name":"School of Integrated Circuits, BNRist, Tsinghua University,Beijing,China,100084"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yang","family":"Hu","sequence":"additional","affiliation":[{"name":"School of Integrated Circuits, BNRist, Tsinghua University,Beijing,China,100084"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shouyi","family":"Yin","sequence":"additional","affiliation":[{"name":"School of Integrated Circuits, BNRist, Tsinghua University,Beijing,China,100084"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","volume-title":"Azurepublicdataset"},{"key":"ref2","volume-title":"Integrating and operating hbm2e memory"},{"key":"ref3","volume-title":"Nvidia a100"},{"key":"ref4","volume-title":"Nvidia b200"},{"key":"ref5","volume-title":"Nvidia gb200 nv172: Powering the new era of computing"},{"key":"ref6","volume-title":"Bard, an experiment by google","year":"2023"},{"key":"ref7","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023","journal-title":"arXiv preprint"},{"key":"ref8","article-title":"Towards a human-like open-domain chatbot","author":"Adiwardana","year":"2020","journal-title":"arXiv preprint"},{"key":"ref9","article-title":"Sarathi: Efficient 11 m inference by piggybacking decodes with chunked prefills","author":"Agrawal","year":"2023","journal-title":"arXiv preprint"},{"key":"ref10","volume-title":"EPYC-7003"},{"key":"ref11","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref12","article-title":"Sparks of artificial general intelligence: Early experiments with gpt-4","author":"Bubeck","year":"2023","journal-title":"arXiv preprint"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA57654.2024.00022"},{"key":"ref14","article-title":"Evaluating large language models trained on code","author":"Chen","year":"2021","journal-title":"arXiv preprint"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001177"},{"issue":"240","key":"ref16","first-page":"1","article-title":"Palm: Scaling language modeling with pathways","volume":"24","author":"Chowdhery","year":"2023","journal-title":"Journal of Machine Learning Research"},{"key":"ref17","article-title":"Flashattention-2: Faster attention with better parallelism and work partitioning","author":"Dao","year":"2023","journal-title":"arXiv preprint"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1189"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1007\/978-981-96-8180-8_8"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2012.2185930"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ECTC51687.2025.00051"},{"key":"ref22","article-title":"The llama 3 herd of models","author":"Dubey","year":"2024","journal-title":"arXiv preprint"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3730999"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3037697.3037702"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1016\/j.mssp.2021.106182"},{"key":"ref26","article-title":"Waferllm: A wafer-scale 11 m inference system","author":"He","year":"2025","journal-title":"arXiv preprint"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/3445814.3446762"},{"key":"ref28","article-title":"Deepspeed-fastgen: High-throughput text generation for 11 ms via mii and deepspeed-inference","author":"Holmes","year":"2024","journal-title":"arXiv preprint"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/TED.2017.2737644"},{"key":"ref30","article-title":"Memserve: Context caching for disaggregated lm serving with elastic memory pool","author":"Hu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref31","article-title":"Inference without interference: Disaggregate 11 m inference for mixed downstream workloads","author":"Hu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/MCAS.2024.3349669"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ECTC51909.2023.00174"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ECTC32696.2021.00028"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00050"},{"key":"ref36","volume-title":"Mistral 7b","author":"Jiang","year":"2023"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1145\/3676641.3715996"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO50266.2020.00058"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3352460.3358252"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/TCPMT.2021.3096786"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/TCPMT.2020.2996658"},{"issue":"06","key":"ref43","first-page":"993","article-title":"Research on wafer-scale chip mapping task based on genetic algorithm","volume":"46","author":"LI","year":"2024","journal-title":"Computer Engineering & Science"},{"key":"ref44","first-page":"663","article-title":"Alpaserve: Statistical multiplexing with model parallelism for deep learning serving","volume-title":"17th USENIX Symposium on Operating Systems Design and Implementation, OSDI 23","author":"Li","year":"2023"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/HCS55958.2022.9895479"},{"key":"ref46","article-title":"Deepseek-v3 technical report","author":"Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00014"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC19947.2020.9063103"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640383"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2011"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18074.2021.9586194"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2019.00042"},{"issue":"700W","key":"ref53","first-page":"1","article-title":"Splitwise: Efficient generative llm inference using phase splitting","volume":"400","author":"Patel","year":"2023","journal-title":"Power"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2024.3388896"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.23919\/DATE64628.2025.10992880"},{"key":"ref56","article-title":"Mooncake: A kvcache-centric disaggregated architecture for 11 m serving","author":"Qin","year":"2024","journal-title":"arXiv preprint"},{"key":"ref57","article-title":"Recipes for building an open-domain chatbot","author":"Roller","year":"2020","journal-title":"arXiv preprint"},{"key":"ref58","article-title":"Dynaserve: Unified and elastic execution for dynamic disaggregated 11 m serving","author":"Ruan","year":"2025","journal-title":"arXiv preprint"},{"key":"ref59","article-title":"Bloom: A 176b-parameter open-access multilingual language model","author":"Scao","year":"2023","journal-title":"arXiv preprint"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS.2010.5536970"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2193"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1145\/3352460.3358302"},{"key":"ref63","article-title":"Nexus: Taming throughput-latency tradeoff in 11 m serving via efficient gpu sharing","author":"Shi","year":"2025","journal-title":"arXiv preprint"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2018.00018"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1145\/3368089.3417058"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/HCS55958.2022.9895534"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00083"},{"key":"ref68","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref69","article-title":"Llama2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv preprint"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/ECTC.2016.65"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.23919\/DATE64628.2025.10993016"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD45719.2019.8942127"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.23919\/ICS.2024.3515003"},{"key":"ref74","first-page":"386","article-title":"Spatial-aware orchestration of 11 m attention on waferscale chips","volume-title":"International Symposium on Advanced Parallel Processing Technologies","author":"Wei","year":"2025"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO61859.2024.00068"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/ISPASS57527.2023.00035"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00086"},{"key":"ref78","first-page":"1","article-title":"Gspo: A graph substitution and parallelization joint optimization framework for dnn inference","volume-title":"Proceedings of the 61st ACM\/IEEE Design Automation Conference","author":"Xu","year":"2024"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731101"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1007\/978-981-95-1021-4_3"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731045"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378514"},{"key":"ref83","article-title":"Flashinfer: Efficient and customizable attention engine for 11 m inference serving","author":"Ye","year":"2025","journal-title":"arXiv preprint"},{"key":"ref84","first-page":"521","article-title":"Orca: A distributed serving system for \\{Transformer-Based\\} generative models","volume-title":"16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22)","author":"Yu","year":"2022"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.1145\/3695053.3731016"},{"key":"ref86","doi-asserted-by":"publisher","DOI":"10.1109\/ISVLSI65124.2025.11130221"},{"key":"ref87","first-page":"787","article-title":"Shepherd: Serving dnns in the wild","volume-title":"20th USENIX Symposium on Networked Systems Design and Implementation, NSDI 23","author":"Zhang","year":"2023"},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.1007\/s11664-022-09503-z"},{"key":"ref89","article-title":"Opt: Open pre-trained transformer language models","author":"Zhang","year":"2022","journal-title":"arXiv preprint"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2000"},{"key":"ref91","article-title":"Distserve: Disaggregating prefill and decoding for goodput-optimized large language model serving","author":"Zhong","year":"2024","journal-title":"arXiv preprint"},{"key":"ref92","article-title":"Theseus: Towards highefficiency wafer-scale chip design space exploration for large language models","author":"Zhu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref93","first-page":"749","article-title":"\\{NanoFlow\\}: Towards optimal large language model serving throughput","volume-title":"19th USENIX Symposium on Operating Systems Design and Implementation (OSDI 25)","author":"Zhu","year":"2025"}],"event":{"name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","location":"Sydney, Australia","start":{"date-parts":[[2026,1,31]]},"end":{"date-parts":[[2026,2,4]]}},"container-title":["2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11408404\/11408433\/11408518.pdf?arnumber=11408518","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T06:35:49Z","timestamp":1772692549000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11408518\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,31]]},"references-count":93,"URL":"https:\/\/doi.org\/10.1109\/hpca68181.2026.11408518","relation":{},"subject":[],"published":{"date-parts":[[2026,1,31]]}}}