Hierarchical Coded Matrix Multiplication

TL;DR

This paper introduces a hierarchical coding approach for distributed matrix multiplication that leverages work done by slower nodes, resulting in significant performance improvements over traditional methods that ignore stragglers.

Contribution

It proposes a novel hierarchical coding scheme that exploits partial work from stragglers, improving distributed matrix multiplication efficiency in heterogeneous environments.

Findings

Achieves 60% reduction in expected finishing time in simulations.

Realizes 35% speedup on Amazon EC2 cloud.

Outperforms existing straggler-resistant methods by utilizing partial computations.

Abstract

Slow working nodes, known as stragglers, can greatly reduce the speed of distributed computation. Coded matrix multiplication is a recently introduced technique that enables straggler-resistant distributed multiplication of large matrices. A key property is that the finishing time depends only on the work completed by a set of the fastest workers, while the work done by the slowest workers is ignored completely. This paper is motivated by the observation that in real-world commercial cloud computing systems such as Amazon's Elastic Compute Cloud (EC2) the distinction between fast and slow nodes is often a soft one. Thus, if we could also exploit the work completed by stragglers we may realize substantial performance gains. To realize such gains, in this paper we use the idea of hierarchical coding (Ferdinand and Draper, IEEE Int. Symp. Inf. Theory, 2018). We decompose the overall matrix multiplication task into a hierarchy of heterogeneously sized subtasks. The duty to complete each subtask is shared amongst all workers and each subtask is (generally) of a different complexity. The motivation for the hierarchical decomposition is the recognition that more workers will finish the first subtask than the second (or third, forth, etc.). Connecting to error correction coding, earlier subtasks can therefore be designed to be of a higher rate than later subtasks. Through this hierarchical design our scheme exploits the work completed by stragglers, rather than ignoring it, even if that amount is much less than that completed by the fastest workers. We numerically show that our method realizes a 60% improvement in the expected finishing time for a widely studied statistical model of the speed of computation and, on Amazon EC2, the gain is 35%.