A Droplet Approach Based on Raptor Codes for Distributed Computing With Straggling Servers

10/08/2018
by   Albin Severinson, et al.
0

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.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/21/2017

Block-Diagonal and LT Codes for Distributed Computing With Straggling Servers

We propose two coded schemes for the distributed computing problem of mu...
research
11/04/2019

Robust low-delay Streaming PIR using convolutional codes

In this paper we investigate the design of a low-delay robust streaming ...
research
05/13/2020

Improved Computation-Communication Trade-Off for Coded Distributed Computing using Linear Dependence of Intermediate Values

In large scale distributed computing systems, communication overhead is ...
research
05/14/2019

Coded Distributed Tracking

We consider the problem of tracking the state of a process that evolves ...
research
09/17/2020

Password similarity using probabilistic data structures

Passwords should be easy to remember, yet expiration policies mandate th...
research
01/21/2020

Serverless Straggler Mitigation using Local Error-Correcting Codes

Inexpensive cloud services, such as serverless computing, are often vuln...
research
02/07/2021

Coded Computing with Noise

Distributed computation is a framework used to break down a complex comp...

Please sign up or login with your details

Forgot password? Click here to reset