Demystifying Parallel and Distributed Deep Learning: An In-Depth Concurrency Analysis

TL;DR

This paper provides a comprehensive analysis of parallel and distributed deep learning, exploring concurrency types, system architectures, and future directions to enhance DNN training efficiency.

Contribution

It offers an in-depth theoretical and practical overview of parallelization strategies and models for deep neural networks, highlighting trends and future research directions.

Findings

Analysis of concurrency in DNNs from operators to distributed systems

Review of asynchronous stochastic optimization and communication schemes

Identification of future directions for parallelism in deep learning

Abstract

Deep Neural Networks (DNNs) are becoming an important tool in modern computing applications. Accelerating their training is a major challenge and techniques range from distributed algorithms to low-level circuit design. In this survey, we describe the problem from a theoretical perspective, followed by approaches for its parallelization. We present trends in DNN architectures and the resulting implications on parallelization strategies. We then review and model the different types of concurrency in DNNs: from the single operator, through parallelism in network inference and training, to distributed deep learning. We discuss asynchronous stochastic optimization, distributed system architectures, communication schemes, and neural architecture search. Based on those approaches, we extrapolate potential directions for parallelism in deep learning.