RPLB: A Replica Placement Algorithm in Data Grid with
Load Balancing
Kingsy Rajaretnam, Manimegalai Rajkumar, and Ranjith
Venkatesan
Department
of Computer Science and Engineering, Sri Ramakrishna Engineering College, India
Abstract: Data grid is an infrastructure
built based on internet which facilitates sharing and management of geographically
distributed data resources. Data sharing in data grids is enhanced through
dynamic data replication methodologies to reduce access latencies and bandwidth
consumption. Replica placement is to create and place duplicate copies of the
most needed file in beneficial locations in the data grid network. To reduce
the make span i.e., total job execution time, storage consumption and effective
network usage in data grids, a new method for replica placement is introduced. In
this proposed method, all the nodes in the same region are grouped together and
replica is placed in the highest degree and highest frequency node in the
region. The node to place replica should be load balanced in terms of access
and storage. The proposed dynamic Replica Placement algorithm with Load Balancing
(RPLB) is tested using OptorSim simulator, which is developed by European Data
Grid Projects. In this paper, two variants of the proposed algorithm RPLB, namely
RPLBfrequency and RPLBdegree are also presented. The
comparative analysis of all the three proposed algorithms is also presented in
this paper. A Graphical User Interface (GUI)
is designed as an interface to OptorSim to get all values for grid
configuration file, job configuration file and parameters configuration file. Simulation
results reveal that the performance of the proposed methodology is better in
terms of makespan, storage consumption and replication count when compared to the
existing algorithms in the literature.
Keywords: Replica
placement, load balancing, effective network usage, data grid, data replication.
Received June 17, 2013; accepted April 28, 2014