Dynamic Backup Workers for Parallel Machine Learning - Archive ouverte HAL Access content directly
Conference Papers Year :

Dynamic Backup Workers for Parallel Machine Learning

(1) , (1) , (1)
1
Chuan Xu
Giovanni Neglia

Abstract

The most popular framework for parallel training of machine learning models is the (synchronous) parameter server (PS). This paradigm consists of n workers and a stateful PS, which waits for the responses of every worker's computation to proceed to the next iteration. Transient computation slowdowns or transmission delays can intolerably lengthen the time of each iteration. An efficient way to mitigate this problem is to let the PS wait only for the fastest n − b updates, before generating the new parameters. The slowest b workers are called backup workers. The optimal number b of backup workers depends on the cluster configuration and workload, but also (as we show in this paper) on the current stage of the training. We propose DBW, an algorithm that dynamically decides the number of backup workers during the training process to maximize the convergence speed at each iteration. Our experiments show that DBW 1) removes the necessity to tune b by preliminary time-consuming experiments, and 2) makes the training up to a factor 3 faster than the optimal static configuration.
Fichier principal
Vignette du fichier
Dynamic_Backup_Workers_Networking (2).pdf (1.26 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03044393 , version 1 (07-12-2020)

Identifiers

  • HAL Id : hal-03044393 , version 1

Cite

Chuan Xu, Giovanni Neglia, Nicola Sebastianelli. Dynamic Backup Workers for Parallel Machine Learning. IFIP Networking 2020, Jun 2020, Paris / Online, France. ⟨hal-03044393⟩
30 View
41 Download

Share

Gmail Facebook Twitter LinkedIn More