OverSketch: Approximate Matrix Multiplication for the Cloud

11/06/2018

∙

by   Vipul Gupta, et al.

∙

0

∙

share

We propose OverSketch, an approximate algorithm for distributed matrix multiplication in serverless computing. OverSketch leverages ideas from matrix sketching and high-performance computing to enable cost-efficient multiplication that is resilient to faults and straggling nodes pervasive in low-cost serverless architectures. We establish statistical guarantees on the accuracy of OverSketch and empirically validate our results by solving a large-scale linear program using interior-point methods and demonstrate a 34 reduction in compute time on AWS Lambda.