An improved GPU MapReduce framework for data intensive applications

Author

Nitu, Razvan ; Apostol, Elena ; Cristea, Valentin

Author_Institution

Fac. of Autom. Control & Comput., Univ. Politeh. of Bucharest, Bucharest, Romania

fYear

2014

fDate

4-6 Sept. 2014

Firstpage

355

Lastpage

362

Abstract

The MapReduce paradigm is one of the best solutions for implementing distributed applications which perform intensive data processing. In terms of performance regarding this type of applications, MapReduce can be improved by adding GPU capabilities. In this context, the GPU clusters for large scale computing can bring a considerable increase in the efficiency and speedup of data intensive applications. In this article we present a framework for executing MapReduce using GPU programming. We describe several improvements to the concept of GPU MapReduce and we compare our solution with others.

Keywords

data handling; graphics processing units; GPU MapReduce framework; GPU clusters; GPU programming; MapReduce paradigm; data intensive applications; distributed applications; intensive data processing; large scale computing; Graphics processing units; Kernel; Mars; Memory management; Parallel processing; Process control; Vectors; GPU MapReduce; Hadoop; OpenCL; shared memory;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Computer Communication and Processing (ICCP), 2014 IEEE International Conference on

Conference_Location

Cluj Napoca

Print_ISBN

978-1-4799-6568-7

Type

conf

DOI

10.1109/ICCP.2014.6937021

Filename

6937021