Yunqi Gao , Zechao Zhang , Bing Hu , A-Long Jin , Chunming Wu
{"title":"OF-WFBP: A near-optimal communication mechanism for tensor fusion in distributed deep learning","authors":"Yunqi Gao , Zechao Zhang , Bing Hu , A-Long Jin , Chunming Wu","doi":"10.1016/j.parco.2023.103053","DOIUrl":null,"url":null,"abstract":"<div><p><span>The communication bottleneck has severely restricted the scalability of distributed deep learning<span>. Tensor fusion improves the scalability of data parallelism by overlapping computation and communication tasks. However, existing tensor fusion schemes only result in suboptimal training performance. In this paper, we propose an efficient communication mechanism (OF-WFBP) to find the optimal tensor fusion scheme for synchronous data parallelism. We present the mathematical model of OF-WFBP and prove it is an NP-hard problem. We mathematically solve the mathematical model of OF-WFBP in two cases. We propose an improved sparrow search algorithm (GradSSA) to find the near-optimal tensor fusion scheme efficiently in other cases. Experimental results on two different </span></span>GPU clusters show that OF-WFBP achieves up to 1.43x speedup compared to the state-of-the-art tensor fusion mechanisms.</p></div>","PeriodicalId":54642,"journal":{"name":"Parallel Computing","volume":"118 ","pages":"Article 103053"},"PeriodicalIF":2.0000,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Parallel Computing","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0167819123000595","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, THEORY & METHODS","Score":null,"Total":0}
引用次数: 0
Abstract
The communication bottleneck has severely restricted the scalability of distributed deep learning. Tensor fusion improves the scalability of data parallelism by overlapping computation and communication tasks. However, existing tensor fusion schemes only result in suboptimal training performance. In this paper, we propose an efficient communication mechanism (OF-WFBP) to find the optimal tensor fusion scheme for synchronous data parallelism. We present the mathematical model of OF-WFBP and prove it is an NP-hard problem. We mathematically solve the mathematical model of OF-WFBP in two cases. We propose an improved sparrow search algorithm (GradSSA) to find the near-optimal tensor fusion scheme efficiently in other cases. Experimental results on two different GPU clusters show that OF-WFBP achieves up to 1.43x speedup compared to the state-of-the-art tensor fusion mechanisms.
期刊介绍：
Parallel Computing is an international journal presenting the practical use of parallel computer systems, including high performance architecture, system software, programming systems and tools, and applications. Within this context the journal covers all aspects of high-end parallel computing from single homogeneous or heterogenous computing nodes to large-scale multi-node systems.
Parallel Computing features original research work and review articles as well as novel or illustrative accounts of application experience with (and techniques for) the use of parallel computers. We also welcome studies reproducing prior publications that either confirm or disprove prior published results.
Particular technical areas of interest include, but are not limited to:
-System software for parallel computer systems including programming languages (new languages as well as compilation techniques), operating systems (including middleware), and resource management (scheduling and load-balancing).
-Enabling software including debuggers, performance tools, and system and numeric libraries.
-General hardware (architecture) concepts, new technologies enabling the realization of such new concepts, and details of commercially available systems
-Software engineering and productivity as it relates to parallel computing
-Applications (including scientific computing, deep learning, machine learning) or tool case studies demonstrating novel ways to achieve parallelism
-Performance measurement results on state-of-the-art systems
-Approaches to effectively utilize large-scale parallel computing including new algorithms or algorithm analysis with demonstrated relevance to real applications using existing or next generation parallel computer architectures.
-Parallel I/O systems both hardware and software
-Networking technology for support of high-speed computing demonstrating the impact of high-speed computation on parallel applications
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
