Welcome to the InfoSci Platform
Reference Hub5
A Two-Level Fuzzy Value-Based Replica Replacement Algorithm in Data Grids

A Two-Level Fuzzy Value-Based Replica Replacement Algorithm in Data Grids

Nazanin Saadat, Amir Masoud Rahmani
Copyright: © 2016 |Volume: 8 |Issue: 4 |Pages: 22
ISSN: 1938-0259|EISSN: 1938-0267|EISBN13: 9781466690011|DOI: 10.4018/IJGHPC.2016100105
Cite Article Cite Article

MLA

Saadat, Nazanin, and Amir Masoud Rahmani. "A Two-Level Fuzzy Value-Based Replica Replacement Algorithm in Data Grids." IJGHPC vol.8, no.4 2016: pp.78-99. http://doi.org/10.4018/IJGHPC.2016100105

APA

Saadat, N. & Rahmani, A. M. (2016). A Two-Level Fuzzy Value-Based Replica Replacement Algorithm in Data Grids. International Journal of Grid and High Performance Computing (IJGHPC), 8(4), 78-99. http://doi.org/10.4018/IJGHPC.2016100105

Chicago

Saadat, Nazanin, and Amir Masoud Rahmani. "A Two-Level Fuzzy Value-Based Replica Replacement Algorithm in Data Grids," International Journal of Grid and High Performance Computing (IJGHPC) 8, no.4: 78-99. http://doi.org/10.4018/IJGHPC.2016100105

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

One of the challenges of data grid is to access widely distributed data fast and efficiently and providing maximum data availability with minimum latency. Data replication is an efficient way used to address this challenge by replicating and storing replicas, making it possible to access similar data in different locations of the data grid and can shorten the time of getting the files. However, as the number and storage size of grid sites is limited and restricted, an optimized and effective replacement algorithm is needed to improve the efficiency of replication. In this paper, the authors propose a novel two-level replacement algorithm which uses Fuzzy Replica Preserving Value Evaluator System (FRPVES) for evaluating the value of each replica. The algorithm was tested using a grid simulator, OptorSim developed by European Data Grid projects. Results from simulation procedure show that the authors' proposed algorithm has better performance in comparison with other algorithms in terms of job execution time, total number of replications and effective network usage.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.