Gradient Coding with Dynamic Clustering for Straggler-Tolerant Distributed Learning

TL;DR

This paper introduces a dynamic gradient coding scheme with clustering that adapts to worker straggling behavior in distributed learning, significantly reducing iteration times without extra communication costs.

Contribution

It proposes a novel GC-DC scheme that dynamically forms clusters based on past straggler behavior, improving efficiency in distributed gradient descent.

Findings

GC-DC reduces average iteration time significantly.

GC-DC performs well in both homogeneous and heterogeneous worker models.

No additional communication load is introduced by GC-DC.

Abstract

Distributed implementations are crucial in speeding up large scale machine learning applications. Distributed gradient descent (GD) is widely employed to parallelize the learning task by distributing the dataset across multiple workers. A significant performance bottleneck for the per-iteration completion time in distributed synchronous GD is $straggling$ workers. Coded distributed computation techniques have been introduced recently to mitigate stragglers and to speed up GD iterations by assigning redundant computations to workers. In this paper, we consider gradient coding (GC), and propose a novel dynamic GC scheme, which assigns redundant data to workers to acquire the flexibility to dynamically choose from among a set of possible codes depending on the past straggling behavior. In particular, we consider GC with clustering, and regulate the number of stragglers in each cluster by dynamically forming the clusters at each iteration; hence, the proposed scheme is called $GC$ $with$ $dynamic$ $clustering$ (GC-DC). Under a time-correlated straggling behavior, GC-DC gains from adapting to the straggling behavior over time such that, at each iteration, GC-DC aims at distributing the stragglers across clusters as uniformly as possible based on the past straggler behavior. For both homogeneous and heterogeneous worker models, we numerically show that GC-DC provides significant improvements in the average per-iteration completion time without an increase in the communication load compared to the original GC scheme.