pith. sign in

A Unified Coding Framework for Distributed Computing with Straggling Servers

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

We propose a unified coded framework for distributed computing with straggling servers, by introducing a tradeoff between "latency of computation" and "load of communication" for some linear computation tasks. We show that the coded scheme of [1]-[3] that repeats the intermediate computations to create coded multicasting opportunities to reduce communication load, and the coded scheme of [4], [5] that generates redundant intermediate computations to combat against straggling servers can be viewed as special instances of the proposed framework, by considering two extremes of this tradeoff: minimizing either the load of communication or the latency of computation individually. Furthermore, the latency-load tradeoff achieved by the proposed coded framework allows to systematically operate at any point on that tradeoff to perform distributed computing tasks. We also prove an information-theoretic lower bound on the latency-load tradeoff, which is shown to be within a constant multiplicative gap from the achieved tradeoff at the two end points.

fields

cs.IT 1

years

2019 1

verdicts

UNVERDICTED 1

representative citing papers

Coded Distributed Computing: Performance Limits and Code Designs

cs.IT · 2019-06-24 · unverdicted · novelty 6.0

Coded distributed computing execution time equals erasure-channel error probability for linear codes, with explicit expressions for binary random linear codes and asymptotic optimality for binary codes matching any linear code.

citing papers explorer

Showing 1 of 1 citing paper.

  • Coded Distributed Computing: Performance Limits and Code Designs cs.IT · 2019-06-24 · unverdicted · none · ref 6 · internal anchor

    Coded distributed computing execution time equals erasure-channel error probability for linear codes, with explicit expressions for binary random linear codes and asymptotic optimality for binary codes matching any linear code.