A Bi-layered Parallel Training Architecture for Large-Scale Convolutional Neural Networks

• Published in: IEEE transactions on parallel and distributed systems Vol. 30; no. 5; pp. 965 - 976
• Main Authors: Chen, Jianguo, Li, Kenli, Bilal, Kashif, Zhou, Xu, Li, Keqin, Yu, Philip S.
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.05.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Acceleration; Artificial neural networks; bi-layered parallel computing; Big data; Computation; Computational modeling; Computer architecture; Computer networks; convolutional neural networks; Datasets; deep learning; Distributed computing; Distributed processing; Iterative methods; Neural networks; Parallel processing; Synchronism; Task analysis; Task scheduling; Training; Weight; Weightlifting; Workload
• DOI: 10.1109/TPDS.2018.2877359
• ISSN: 1045-9219

Benefitting from large-scale training datasets and the complex training network, Convolutional Neural Networks (CNNs) are widely applied in various fields with high accuracy. However, the training process of CNNs is very time-consuming, where large amounts of training samples and iterative operations are required to obtain high-quality weight parameters. In this paper, we focus on the time-consuming training process of large-scale CNNs and propose a Bi-layered Parallel Training (BPT-CNN) architecture in distributed computing environments. BPT-CNN consists of two main components: (a) an outer-layer parallel training for multiple CNN subnetworks on separate data subsets, and (b) an inner-layer parallel training for each subnetwork. In the outer-layer parallelism, we address critical issues of distributed and parallel computing, including data communication, synchronization, and workload balance. A heterogeneous-aware Incremental Data Partitioning and Allocation (IDPA) strategy is proposed, where large-scale training datasets are partitioned and allocated to the computing nodes in batches according to their computing power. To minimize the synchronization waiting during the global weight update process, an Asynchronous Global Weight Update (AGWU) strategy is proposed. In the inner-layer parallelism, we further accelerate the training process for each CNN subnetwork on each computer, where computation steps of convolutional layer and the local weight training are parallelized based on task-parallelism. We introduce task decomposition and scheduling strategies with the objectives of thread-level load balancing and minimum waiting time for critical paths. Extensive experimental results indicate that the proposed BPT-CNN effectively improves the training performance of CNNs while maintaining the accuracy.