TPC-H benchmarking of Pig Latin on a Hadoop cluster

Author

Moussa, Rim

Author_Institution

Lab. ITC&E Eng., Univ. of Tunis, Tunis, Tunisia

fYear

2012

fDate

26-28 June 2012

Firstpage

85

Lastpage

90

Abstract

Several companies report success stories after migration from relational database management systems to NoSQL systems (Not only SQL). The latter seem to take over in most data storage fields. These technologies must be used properly, and businesses must be aware of the limitations of NoSQL, for providing real benefits. Pig Latin is a high-level language for expressing data analysis programs, and implementing the MapReduce framework on top of Hadoop Distributed File System. This paper benchmarks Pig Latin using the well known TPC-H benchmark -a Decision Support System benchmark, and reports performance results for different settings on GRID5000 clusters.

Keywords

SQL; data analysis; decision support systems; distributed processing; high level languages; pattern clustering; relational databases; GRID5000 clusters; Hadoop cluster; Hadoop distributed file system; MapReduce framework; NoSQL systems; Pig Latin; TPC-H benchmarking; data analysis programs; data storage fields; decision support system benchmark; high-level language; relational database management systems; Benchmark testing; Business; Engines; High level languages; Parallel processing; Relational databases; Time factors; Hadoop; MapReduce; Pig Latin; TPC-H; analytics; benchmark; cloud;

fLanguage

English

Publisher

ieee

Conference_Titel

Communications and Information Technology (ICCIT), 2012 International Conference on

Conference_Location

Hammamet

Print_ISBN

978-1-4673-1949-2

Type

conf

DOI

10.1109/ICCITechnol.2012.6285848

Filename

6285848