{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T05:45:06Z","timestamp":1782798306337,"version":"3.54.5"},"reference-count":51,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T00:00:00Z","timestamp":1779062400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T00:00:00Z","timestamp":1779062400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,5,18]]},"DOI":"10.1109\/infocom59046.2026.11571473","type":"proceedings-article","created":{"date-parts":[[2026,6,29]],"date-time":"2026-06-29T19:38:15Z","timestamp":1782761895000},"page":"1-10","source":"Crossref","is-referenced-by-count":0,"title":["xCCLTuner: Treating xCCL as Black-Box and Automatically Tuning"],"prefix":"10.1109","author":[{"given":"Chenxu","family":"Wang","sequence":"first","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wentao","family":"Fan","sequence":"additional","affiliation":[{"name":"China Mobile (Suzhou) Software Technology Co. Ltd.,Suzhou,China,215000"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhehao","family":"Lin","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peirui","family":"Cao","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xi","family":"Lin","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaohu","family":"Xu","sequence":"additional","affiliation":[{"name":"China Mobile (Suzhou) Software Technology Co. Ltd.,Suzhou,China,215000"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wanchun","family":"Dou","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guihai","family":"Chen","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chen","family":"Tian","sequence":"additional","affiliation":[{"name":"Nanjing University,State Key Laboratory for Novel Software Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/s11390-023-2894-6"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CCGrid51090.2021.00049"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1145\/3452296.3472904"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/j.parco.2009.09.001"},{"key":"ref5","article-title":"Highly scalable deep learning training system with mixed-precision: Training imagenet in four minutes. corr abs\/1807.11205 (2018)","author":"Jia","year":"2018"},{"key":"ref6","first-page":"770","article-title":"Imagenet\/resnet-50 training in 224 seconds","author":"Mikami","year":"2018"},{"key":"ref7","article-title":"Image classification at supercomputer scale","author":"Ying","year":"2018"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651362"},{"key":"ref9","first-page":"593","article-title":"{TACCL}: Guiding collective algorithm synthesis using communication sketches","volume-title":"NSDI","author":"Shah"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3437801.3441620"},{"key":"ref11","first-page":"1445","article-title":"Swing: Short-cutting rings for higher bandwidth allreduce","volume-title":"NSDI","author":"De Sensi"},{"key":"ref12","first-page":"745","article-title":"{MegaScale}: Scaling large language model training to more than 10,000 {GPUs}","volume-title":"NSDI","author":"Jiang"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3651890.3672265"},{"key":"ref14","article-title":"Imagenet classification with deep convolutional neural networks","volume-title":"NIPS","volume":"25","author":"Krizhevsky"},{"key":"ref15","article-title":"Megatron-lm: Training multi-billion parameter language models using model parallelism","author":"Shoeybi","year":"2019"},{"key":"ref16","article-title":"Deep learning recommendation model for personalization and recommendation systems","volume-title":"CoRR","volume":"abs\/1906.00091","author":"Naumov","year":"2019"},{"key":"ref17","article-title":"Rocm communication collectives library","year":"2024"},{"key":"ref18","article-title":"oneccl","year":"2024"},{"key":"ref19","article-title":"Microsoft collective communication library (msccl)","year":"2021"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2021.3091475"},{"issue":"54","key":"ref21","first-page":"1","article-title":"Smac3: A versatile bayesian optimization package for hyperparameter optimization","volume":"23","author":"Lindauer","year":"2022","journal-title":"J. Mach. Learn. Res"},{"key":"ref22","first-page":"561","article-title":"Ray: a distributed framework for emerging ai applications","volume-title":"OSDI","author":"Moritz"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3406703"},{"key":"ref24","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume-title":"NIPS","author":"Paszke"},{"key":"ref25","article-title":"Efficient estimation of word representations in vector space","author":"Mikolov","year":"2013"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1162"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/3545611"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1002\/widm.1484"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1613\/jair.4806"},{"key":"ref30","first-page":"493","article-title":"High-dimensional bayesian optimization with sparse axis-aligned subspaces","volume-title":"UAI","author":"Eriksson"},{"key":"ref31","first-page":"4752","article-title":"A framework for bayesian optimization in embedded subspaces","volume-title":"ICML","author":"Nayebi"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.14778\/3632093.3632114"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1016\/B978-1-55860-335-6.50026-X"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1163\/156853974X00534"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.5555\/2188385.2188395"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref37","first-page":"418","article-title":"Tictac: Accelerating distributed deep learning with communication scheduling","volume":"1","author":"Hashemi","year":"2019","journal-title":"MLSys"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1145\/3477132.3483553"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.14778\/3681954.3681968"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2023.3266246"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/3503222.3507778"},{"key":"ref42","article-title":"Flux: Fast software-based communication overlap on gpus through kernel fusion","author":"Chang","year":"2024"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1145\/3567955.3567959"},{"key":"ref44","first-page":"809","article-title":"Better together: Jointly optimizing {ML} collective scheduling and execution planning using {SYNDICATE}","volume-title":"NSDI","author":"Mahajan"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00704"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1145\/3650200.3656636"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/BigData52589.2021.9671389"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/DLS49591.2019.00016"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPSW50202.2020.00171"},{"key":"ref50","article-title":"torchrec: Super slow allreduce in multi-node multi-gpu setting"},{"key":"ref51","article-title":"Horovod: fast and easy distributed deep learning in tensorflow","author":"Sergeev","year":"2018"}],"event":{"name":"IEEE INFOCOM 2026 - IEEE Conference on Computer Communications","location":"Tokyo, Japan","start":{"date-parts":[[2026,5,18]]},"end":{"date-parts":[[2026,5,21]]}},"container-title":["IEEE INFOCOM 2026 - IEEE Conference on Computer Communications"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11571071\/11571169\/11571473.pdf?arnumber=11571473","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T05:19:32Z","timestamp":1782796772000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11571473\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,18]]},"references-count":51,"URL":"https:\/\/doi.org\/10.1109\/infocom59046.2026.11571473","relation":{},"subject":[],"published":{"date-parts":[[2026,5,18]]}}}