A Droplet Approach Based on Raptor Codes for Distributed Computing With Straggling Servers
Information Theory
2018-10-09 v1 Distributed, Parallel, and Cluster Computing
Machine Learning
Performance
math.IT
Abstract
We propose a coded distributed computing scheme based on Raptor codes to address the straggler problem. In particular, we consider a scheme where each server computes intermediate values, referred to as droplets, that are either stored locally or sent over the network. Once enough droplets are collected, the computation can be completed. Compared to previous schemes in the literature, our proposed scheme achieves lower computational delay when the decoding time is taken into account.
Cite
@article{arxiv.1810.03488,
title = {A Droplet Approach Based on Raptor Codes for Distributed Computing With Straggling Servers},
author = {Albin Severinson and Alexandre Graell i Amat and Eirik Rosnes and Francisco Lazaro and Gianluigi Liva},
journal= {arXiv preprint arXiv:1810.03488},
year = {2018}
}
Comments
Accepted at the 2018 International Symposium on Turbo Codes & Iterative Information Processing, Hong Kong