Shengwei Li;Kai Lu;Zhiquan Lai;Weijie Liu;Keshi Ge;Dongsheng Li
{"title":"A Multidimensional Communication Scheduling Method for Hybrid Parallel DNN Training","authors":"Shengwei Li;Kai Lu;Zhiquan Lai;Weijie Liu;Keshi Ge;Dongsheng Li","doi":"10.1109/TPDS.2024.3406420","DOIUrl":null,"url":null,"abstract":"The transformer-based deep neural network (DNN) models have shown considerable success across diverse tasks, prompting widespread adoption of distributed training methods such as data parallelism and pipeline parallelism. With the increasing parameter number, hybrid parallel training becomes imperative to scale training. The primary bottleneck in scaling remains the communication overhead. The communication scheduling technique, emphasizing the overlap of communication with computation, has demonstrated its benefits in scaling. However, most existing works focus on data parallelism, overlooking the nuances of hybrid parallel training. In this paper, we propose \n<monospace>TriRace</monospace>\n, an efficient communication scheduling framework for accelerating communications in hybrid parallel training of asynchronous pipeline parallelism and data parallelism. To achieve effective computation-communication overlap, \n<monospace>TriRace</monospace>\n introduces \n<italic>3D communication scheduling</i>\n, which adeptly leverages data dependencies between communication and computations, efficiently scheduling AllReduce communication, sparse communication, and peer-to-peer communication in hybrid parallel training. To avoid possible communication contentions, \n<monospace>TriRace</monospace>\n also incorporates a \n<italic>topology-aware runtime</i>\n which optimizes the execution of communication operations by considering ongoing communication operations and real-time network status. We have implemented a prototype of \n<monospace>TriRace</monospace>\n based on PyTorch and Pipedream-2BW, and conducted comprehensive evaluations with three representative baselines. Experimental results show that \n<monospace>TriRace</monospace>\n achieves up to 1.07–1.45× speedup compared to the state-of-the-art pipeline parallelism training baseline Pipedream-2BW, and 1.24–1.81× speedup compared to the Megatron.","PeriodicalId":13257,"journal":{"name":"IEEE Transactions on Parallel and Distributed Systems","volume":null,"pages":null},"PeriodicalIF":5.6000,"publicationDate":"2024-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Parallel and Distributed Systems","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/10540313/","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, THEORY & METHODS","Score":null,"Total":0}
引用次数: 0
Abstract
The transformer-based deep neural network (DNN) models have shown considerable success across diverse tasks, prompting widespread adoption of distributed training methods such as data parallelism and pipeline parallelism. With the increasing parameter number, hybrid parallel training becomes imperative to scale training. The primary bottleneck in scaling remains the communication overhead. The communication scheduling technique, emphasizing the overlap of communication with computation, has demonstrated its benefits in scaling. However, most existing works focus on data parallelism, overlooking the nuances of hybrid parallel training. In this paper, we propose
TriRace
, an efficient communication scheduling framework for accelerating communications in hybrid parallel training of asynchronous pipeline parallelism and data parallelism. To achieve effective computation-communication overlap,
TriRace
introduces
3D communication scheduling
, which adeptly leverages data dependencies between communication and computations, efficiently scheduling AllReduce communication, sparse communication, and peer-to-peer communication in hybrid parallel training. To avoid possible communication contentions,
TriRace
also incorporates a
topology-aware runtime
which optimizes the execution of communication operations by considering ongoing communication operations and real-time network status. We have implemented a prototype of
TriRace
based on PyTorch and Pipedream-2BW, and conducted comprehensive evaluations with three representative baselines. Experimental results show that
TriRace
achieves up to 1.07–1.45× speedup compared to the state-of-the-art pipeline parallelism training baseline Pipedream-2BW, and 1.24–1.81× speedup compared to the Megatron.
期刊介绍:
IEEE Transactions on Parallel and Distributed Systems (TPDS) is published monthly. It publishes a range of papers, comments on previously published papers, and survey articles that deal with the parallel and distributed systems research areas of current importance to our readers. Particular areas of interest include, but are not limited to:
a) Parallel and distributed algorithms, focusing on topics such as: models of computation; numerical, combinatorial, and data-intensive parallel algorithms, scalability of algorithms and data structures for parallel and distributed systems, communication and synchronization protocols, network algorithms, scheduling, and load balancing.
b) Applications of parallel and distributed computing, including computational and data-enabled science and engineering, big data applications, parallel crowd sourcing, large-scale social network analysis, management of big data, cloud and grid computing, scientific and biomedical applications, mobile computing, and cyber-physical systems.
c) Parallel and distributed architectures, including architectures for instruction-level and thread-level parallelism; design, analysis, implementation, fault resilience and performance measurements of multiple-processor systems; multicore processors, heterogeneous many-core systems; petascale and exascale systems designs; novel big data architectures; special purpose architectures, including graphics processors, signal processors, network processors, media accelerators, and other special purpose processors and accelerators; impact of technology on architecture; network and interconnect architectures; parallel I/O and storage systems; architecture of the memory hierarchy; power-efficient and green computing architectures; dependable architectures; and performance modeling and evaluation.
d) Parallel and distributed software, including parallel and multicore programming languages and compilers, runtime systems, operating systems, Internet computing and web services, resource management including green computing, middleware for grids, clouds, and data centers, libraries, performance modeling and evaluation, parallel programming paradigms, and programming environments and tools.