Preserving Sparsity and Privacy in Straggler-Resilient Distributed Matrix Computations

TL;DR

This paper introduces a new distributed matrix computation scheme that preserves input sparsity, optimizes coding weight for resilience to stragglers, and balances privacy and computation time, validated through AWS experiments.

Contribution

It presents a scheme that achieves the lower bound on coding weight for straggler resilience while maintaining input sparsity and privacy in distributed matrix computations.

Findings

Achieves optimal coding weight for maximum straggler resilience.

Preserves input sparsity to enhance computational efficiency.

Validated improvements in straggler mitigation and speed on AWS.

Abstract

Existing approaches to distributed matrix computations involve allocating coded combinations of submatrices to worker nodes, to build resilience to stragglers and/or enhance privacy. In this study, we consider the challenge of preserving input sparsity in such approaches to retain the associated computational efficiency enhancements. First, we find a lower bound on the weight of coding, i.e., the number of submatrices to be combined to obtain coded submatrices to provide the resilience to the maximum possible number of stragglers (for given number of nodes and their storage constraints). Next we propose a distributed matrix computation scheme which meets this exact lower bound on the weight of the coding. Further, we develop controllable trade-off between worker computation time and the privacy constraint for sparse input matrices in settings where the worker nodes are honest but curious. Numerical experiments conducted in Amazon Web Services (AWS) validate our assertions regarding straggler mitigation and computation speed for sparse matrices.