Data Partitioning for Minimizing Transferred Data in MapReduce - LIRMM - Laboratoire d’Informatique, de Robotique et de Microélectronique de Montpellier
Conference Papers Year : 2013

Data Partitioning for Minimizing Transferred Data in MapReduce

Miguel Liroz-Gistau
  • Function : Author
  • PersonId : 901689
Reza Akbarinia
Esther Pacitti

Abstract

Reducing data transfer in MapReduce's shuffle phase is very important because it increases data locality of reduce tasks, and thus decreases the overhead of job executions. In the literature, several optimizations have been proposed to reduce data transfer between mappers and reducers. Nevertheless, all these approaches are limited by how intermediate key-value pairs are distributed over map outputs. In this paper, we address the problem of high data transfers in MapReduce, and propose a technique that repartitions tuples of the input datasets, and thereby optimizes the distribution of key-values over mappers, and increases the data locality in reduce tasks. Our approach captures the relationships between input tuples and intermediate keys by monitoring the execution of a set of MapReduce jobs which are representative of the workload. Then, based on those relationships, it assigns input tuples to the appropriate chunks. We evaluated our approach through experimentation in a Hadoop deployment on top of Grid5000 using standard benchmarks. The results show high reduction in data transfer during the shuffle phase compared to Native Hadoop.
Fichier principal
Vignette du fichier
globe_2013-paper.pdf (194.91 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

lirmm-00879527 , version 1 (04-11-2013)

Identifiers

Cite

Miguel Liroz-Gistau, Reza Akbarinia, Divyakant Agrawal, Esther Pacitti, Patrick Valduriez. Data Partitioning for Minimizing Transferred Data in MapReduce. Globe, Aug 2013, Prague, Czech Republic. pp.1-12, ⟨10.1007/978-3-642-40053-7_1⟩. ⟨lirmm-00879527⟩
659 View
1091 Download

Altmetric

Share

More