{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,24]],"date-time":"2026-02-24T17:59:26Z","timestamp":1771955966786,"version":"3.50.1"},"publisher-location":"New York, NY, USA","reference-count":29,"publisher":"ACM","license":[{"start":{"date-parts":[[2022,8,29]],"date-time":"2022-08-29T00:00:00Z","timestamp":1661731200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U20A20226"],"award-info":[{"award-number":["U20A20226"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2022,8,29]]},"DOI":"10.1145\/3545008.3545031","type":"proceedings-article","created":{"date-parts":[[2023,1,15]],"date-time":"2023-01-15T01:04:08Z","timestamp":1673744648000},"page":"1-12","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":5,"title":["Automatically Generating High-performance Matrix Multiplication Kernels on the Latest Sunway Processor"],"prefix":"10.1145","author":[{"given":"Xiaohan","family":"Tao","sequence":"first","affiliation":[{"name":"State Key Laboratory of Mathematical Engineering and Advanced Computing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yu","family":"Zhu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Mathematical Engineering and Advanced Computing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Boyang","family":"Wang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Mathematical Engineering and Advanced Computing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jinlong","family":"Xu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Mathematical Engineering and Advanced Computing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jianmin","family":"Pang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Mathematical Engineering and Advanced Computing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jie","family":"Zhao","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Mathematical Engineering and Advanced Computing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2023,1,13]]},"reference":[{"key":"e_1_3_2_1_1_1","unstructured":"Somashekaracharya\u00a0G. Bhaskaracharya Julien Demouth and Vinod Grover. 2020. Automatic Kernel Generation for Volta Tensor Cores. arxiv:2006.12645\u00a0[cs.PL]"},{"key":"e_1_3_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1145\/263580.263662"},{"key":"e_1_3_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1145\/2503210.2503289"},{"key":"e_1_3_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/1375581.1375595"},{"key":"e_1_3_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1145\/3126908.3126910"},{"key":"e_1_3_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-016-5588-7"},{"key":"e_1_3_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/3337821.3337883"},{"key":"e_1_3_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1145\/1356052.1356053"},{"key":"e_1_3_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/2743016"},{"key":"e_1_3_2_1_10_1","unstructured":"National Supercomputing\u00a0Center in Wuxi. 2016. xMath User Manual v1.0\u00a0(in Chinese). http:\/\/www.nsccwx.cn:1337\/uploads\/595bce0bed1b4537994d927ef6be922d.pdf"},{"key":"e_1_3_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICPP.2017.51"},{"key":"e_1_3_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.1145\/3378176"},{"key":"e_1_3_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1145\/3497776.3517770"},{"key":"e_1_3_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICAPP.1995.472180"},{"key":"e_1_3_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/2491956.2462187"},{"key":"e_1_3_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1145\/2925987"},{"key":"e_1_3_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.5555\/3049832.3049846"},{"key":"e_1_3_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3433103"},{"key":"e_1_3_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/2764454"},{"key":"e_1_3_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.1145\/3355606"},{"key":"e_1_3_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-15582-6_49"},{"key":"e_1_3_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/2400682.2400713"},{"key":"e_1_3_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/2503210.2503219"},{"key":"e_1_3_2_1_24_1","volume-title":"Proceedings of the 1998 ACM\/IEEE Conference on Supercomputing","author":"R.","unstructured":"R.\u00a0Clint Whaley and Jack\u00a0J. Dongarra. 1998. Automatically Tuned Linear Algebra Software. In Proceedings of the 1998 ACM\/IEEE Conference on Supercomputing (San Jose, CA) (SC\u201998). IEEE Computer Society, USA, 1\u201327."},{"key":"e_1_3_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2018.00086"},{"key":"e_1_3_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2014.14"},{"key":"e_1_3_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO50266.2020.00044"},{"key":"e_1_3_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/3453483.3454106"},{"key":"e_1_3_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1145\/3458817.3476158"}],"event":{"name":"ICPP '22: 51st International Conference on Parallel Processing","location":"Bordeaux France","acronym":"ICPP '22"},"container-title":["Proceedings of the 51st International Conference on Parallel Processing"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3545008.3545031","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3545008.3545031","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T19:02:43Z","timestamp":1750186963000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3545008.3545031"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,29]]},"references-count":29,"alternative-id":["10.1145\/3545008.3545031","10.1145\/3545008"],"URL":"https:\/\/doi.org\/10.1145\/3545008.3545031","relation":{},"subject":[],"published":{"date-parts":[[2022,8,29]]},"assertion":[{"value":"2023-01-13","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}