Wei Wen, Quanyu Zhu, Weiwei Chu, Wen-Yen Chen, Jiyan Yang
{"title":"CubicML:分布式 ML 系统的自动 ML 协同设计与 ML 性能预测","authors":"Wei Wen, Quanyu Zhu, Weiwei Chu, Wen-Yen Chen, Jiyan Yang","doi":"arxiv-2409.04585","DOIUrl":null,"url":null,"abstract":"Scaling up deep learning models has been proven effective to improve\nintelligence of machine learning (ML) models, especially for industry\nrecommendation models and large language models. The co-design of distributed\nML systems and algorithms (to maximize training performance) plays a pivotal\nrole for its success. As it scales, the number of co-design hyper-parameters\ngrows rapidly which brings challenges to feasibly find the optimal setup for\nsystem performance maximization. In this paper, we propose CubicML which uses\nML to automatically optimize training performance of distributed ML systems. In\nCubicML, we use a ML model as a proxy to predict the training performance for\nsearch efficiency and performance modeling flexibility. We proved that CubicML\ncan effectively optimize training speed of in-house ads recommendation models\nand large language models at Meta.","PeriodicalId":501422,"journal":{"name":"arXiv - CS - Distributed, Parallel, and Cluster Computing","volume":"75 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"CubicML: Automated ML for Distributed ML Systems Co-design with ML Prediction of Performance\",\"authors\":\"Wei Wen, Quanyu Zhu, Weiwei Chu, Wen-Yen Chen, Jiyan Yang\",\"doi\":\"arxiv-2409.04585\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Scaling up deep learning models has been proven effective to improve\\nintelligence of machine learning (ML) models, especially for industry\\nrecommendation models and large language models. The co-design of distributed\\nML systems and algorithms (to maximize training performance) plays a pivotal\\nrole for its success. As it scales, the number of co-design hyper-parameters\\ngrows rapidly which brings challenges to feasibly find the optimal setup for\\nsystem performance maximization. In this paper, we propose CubicML which uses\\nML to automatically optimize training performance of distributed ML systems. In\\nCubicML, we use a ML model as a proxy to predict the training performance for\\nsearch efficiency and performance modeling flexibility. We proved that CubicML\\ncan effectively optimize training speed of in-house ads recommendation models\\nand large language models at Meta.\",\"PeriodicalId\":501422,\"journal\":{\"name\":\"arXiv - CS - Distributed, Parallel, and Cluster Computing\",\"volume\":\"75 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Distributed, Parallel, and Cluster Computing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.04585\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Distributed, Parallel, and Cluster Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.04585","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
扩展深度学习模型已被证明能有效提高机器学习(ML)模型的智能,特别是对于行业推荐模型和大型语言模型。分布式 ML 系统和算法的协同设计(以最大限度地提高训练性能)对其成功起着关键作用。随着系统规模的扩大,协同设计超参数的数量也在迅速增加,这给找到系统性能最大化的最佳设置带来了挑战。在本文中,我们提出了 CubicML,它使用ML 自动优化分布式 ML 系统的训练性能。在 CubicML 中,我们使用一个 ML 模型作为代理来预测训练性能,以提高搜索效率和性能建模的灵活性。我们证明,CubicML 可以有效优化 Meta 公司内部广告推荐模型和大型语言模型的训练速度。
CubicML: Automated ML for Distributed ML Systems Co-design with ML Prediction of Performance
Scaling up deep learning models has been proven effective to improve
intelligence of machine learning (ML) models, especially for industry
recommendation models and large language models. The co-design of distributed
ML systems and algorithms (to maximize training performance) plays a pivotal
role for its success. As it scales, the number of co-design hyper-parameters
grows rapidly which brings challenges to feasibly find the optimal setup for
system performance maximization. In this paper, we propose CubicML which uses
ML to automatically optimize training performance of distributed ML systems. In
CubicML, we use a ML model as a proxy to predict the training performance for
search efficiency and performance modeling flexibility. We proved that CubicML
can effectively optimize training speed of in-house ads recommendation models
and large language models at Meta.