INTERNATIONAL JOURNAL OF ENHANCED RESEARCH IN SCIENCE TECHNOLOGY & ENGINEERING VOL. 2 ISSUE 3, MARCH-2013 ISSN NO: 2319-7463 www.erpublications.com 1 oDASuANCO - Ant Colony Optimization based Data Allocation Strategy in Peer-to-Peer Distributed Databases Dr. D.I. George Amalarethinam 1 , C. Balakrishnan 2 1 Director-MCA, Associate Professor, Department of Computer Science, Jamal Mohamed College (Autonomous), Tiruchirappalli, India 2 Assistant Professor, Department of Computer Science, St. Joseph’s College (Autonomous), Tiruchirappalli, India 1 di_geogre@jmc.edu 2 balasjc@gmail.com Abstract: Data Allocation Problem (DAP) incorporates Allocation and Replication methodologies which are involved during the phase of Distributed Database design. Accessibility and availability are the thriving factors for better design of Distributed Databases. High degree of Accessibility and Availability are the outcomes of effective methodologies for Allocation and Replication. This paper proposes a methodology for Reallocation and Replication of fragments in Peer-to-Peer Distributed Database Systems (P2PDDBS) environment using Ant Colony Optimization (ACO) principle, namely, optimized Data Allocation Strategy using Ant Colony Optimization (oDASuANCO). ACO algorithm is a meta-heuristic, works with swarm intelligence technique. The experimental results show that ACO based reallocation and replication methodology improves the ratio of successful queries than the initial allocation of fragments. Keywords: Ant Colony Optimization algorithm, Data Allocation Problem, FlexiPeer, Peer-to-Peer Distributed Databases, Reallocation, Replication 1. Introduction The Distributed Database Systems (DDBS) are more compatible not only with decentralized nature and ability to store growing volume of data, but also to serve the queries in an effective manner by providing a higher degree of parallelism, and improved availability as well as accessibility [1]. Accessibility and availability are the most important aspects to be addressed by any of DDBS environment [2]. These aspects improves the QoS factors like fault tolerance, query execution time and possibilities for recovery. Accessibility is achieved by means of better allocation scheme and better replication scheme leading to accomplish high degree of availability. Particularly, this research considers Reallocation of fragments to appropriate places. Reallocation can be considered as an extended approach of allocation. Reallocation can tune-up the process of allocation and enhances the performance. Reallocation can reduce the drawbacks of allocation and increases the success ratio by allocating the fragments into appropriate sites, for which, reallocation requires the performance analysis information of initial allocation. Hence, the number of frequent transactions tried to access the particular fragment from the particular site is considered as the performance analysis information of initial allocation. The main objective of this paper is to propose a methodology using Ant Colony Optimization (ACO) for deriving Data Allocation Strategy focusing on reallocation and replication, namely, 'optimized Data Allocation Strategy using Ant Colony Optimization' (oDASuANCO). The Simulation result justifies the effectiveness of oDASuANCO based reallocation and replication that the proposed methodology increases the number of successful transactions than initial allocation. The next section narrates the related works on reallocation and replication management. The section III describes the initial allocation. The section IV describes the criteria for using heuristic algorithm like ACO to solve DAP. Section V explains the methodology of oDASuANCO. Section VI elaborates the simulation results of oDASuANCO. The section VII concludes the work. 2. Literature Review This section of the paper states the related works that are stimulated to do research on reallocation and replication methodology. Allocation is the next phase that follows fragmentation in DDBS design. Better allocation schemes result in high degree of accessibility. The following section describes the heuristic methodologies used for allocation and reallocation of fragments to the sites in DDBS Rosa Karimi Adl et al., [1] coined a heuristic algorithm for fragment allocation which is based on the ant colony optimization, a meta-heuristic algorithm. The ACO based approach applied for query optimization and integrity enforcement in DDBS. The goal is for efficient data allocation scheme and to minimize the total transaction response time under memory capacity constraints of the sites. Yin-Fu Huang and Jyh-Her Chen [3] proposed a simple and comprehensive model for a fragment allocation problem, also developed two heuristics algorithms to find an optimal allocation of the fragments. Reza Basseda et al. [4] introduced the fuzzy inference engine to the existing Near Neighborhood Allocation (NNA) algorithm and studied the performance improvement of new Fuzzy Neighborhood Allocation method. Adrian Runceanu [5] drafted an evaluation tool, namely, EvalTool to implement a heuristic algorithm for vertical fragmentation.