DISCO: Distributed Inference with Sparse Communications

TL;DR

DISCO introduces a distributed inference framework that reduces communication, computation, and memory requirements for large DNNs by optimizing sparse data exchanges across multiple nodes, enabling efficient deployment on resource-limited devices.

Contribution

The paper proposes a novel training framework for sparse communication in distributed DNN inference, optimizing data transmission to significantly reduce communication and computation costs.

Findings

Achieves 5x less data communication in ResNet-50 inference.

Reduces overall computation and memory by nearly half.

Provides 4.7x speedup in inference speed.

Abstract

Deep neural networks (DNNs) have great potential to solve many real-world problems, but they usually require an extensive amount of computation and memory. It is of great difficulty to deploy a large DNN model to a single resource-limited device with small memory capacity. Distributed computing is a common approach to reduce single-node memory consumption and to accelerate the inference of DNN models. In this paper, we explore the "within-layer model parallelism", which distributes the inference of each layer into multiple nodes. In this way, the memory requirement can be distributed to many nodes, making it possible to use several edge devices to infer a large DNN model. Due to the dependency within each layer, data communications between nodes during this parallel inference can be a bottleneck when the communication bandwidth is limited. We propose a framework to train DNN models for Distributed Inference with Sparse Communications (DISCO). We convert the problem of selecting which subset of data to transmit between nodes into a model optimization problem, and derive models with both computation and communication reduction when each layer is inferred on multiple nodes. We show the benefit of the DISCO framework on a variety of CV tasks such as image classification, object detection, semantic segmentation, and image super resolution. The corresponding models include important DNN building blocks such as convolutions and transformers. For example, each layer of a ResNet-50 model can be distributively inferred across two nodes with five times less data communications, almost half overall computations and half memory requirement for a single node, and achieve comparable accuracy to the original ResNet-50 model. This also results in 4.7 times overall inference speedup.