Distributed ACS Algorithm for Resource Discovery in Grid

Grid is an environment that allows sharing of resources that are managed by diverse, independent administrative organizations that are geographically distributed. The main objective of grid is to enable users to solve problems using the available geographically distributed resources. To fully utilize these resources, effective discovery techniques are necessities. Grid resource discovery refers to the process of locating satisfactory resources based on user requests. In this paper, we propose a mobile agent method based on peer to peer model for the resource discovery problem that has essential characteristics for efficient, self-configuring and faulttolerant resource discovery and is able to handle dynamic attributes. This method employs an Ant Colony System (ACS) algorithm to locate the required resources. The innovation in this algorithm is to eliminate centralized control and provide node autonomy. We compare our approach with flooding algorithm and Ant System algorithm in terms of Average Hop count, Average Success rate and Querying traffic. Our experiments show that the proposed method performs better than flooding algorithm and Ant System algorithm alike. Keywords— Resource Discovery; Grid; Ant Colony System Algorithm

[1]  Marco Dorigo,et al.  Distributed Optimization by Ant Colonies , 1992 .

[2]  Ian Foster,et al.  On Fully Decentralized Resource Discovery in Grid Environments , 2001, GRID.

[3]  Albert-László Barabási,et al.  Scale-free networks , 2008, Scholarpedia.

[4]  Domenico Talia,et al.  Peer-to-Peer Models for Resource Discovery in Large-Scale Grids: A Scalable Architecture , 2006, VECPAR.

[5]  Albert,et al.  Emergence of scaling in random networks , 1999, Science.

[6]  Junwei Cao,et al.  Self-organizing agents for grid load balancing , 2004, Fifth IEEE/ACM International Workshop on Grid Computing.

[7]  S. Wicker,et al.  Termite: ad-hoc networking with stigmergy , 2003, GLOBECOM '03. IEEE Global Telecommunications Conference (IEEE Cat. No.03CH37489).

[8]  Artur Andrzejak,et al.  Scalable, efficient range queries for grid information services , 2002, Proceedings. Second International Conference on Peer-to-Peer Computing,.

[9]  John Levine,et al.  A hybrid ant algorithm for scheduling independent jobs in heterogeneous computing environments , 2004 .

[10]  Miron Livny,et al.  Experience with the Condor distributed batch system , 1990, IEEE Workshop on Experimental Distributed Systems.

[11]  Min Cai,et al.  MAAN: A Multi-Attribute Addressable Network for Grid Information Services , 2003, Journal of Grid Computing.

[12]  Yuhui Deng,et al.  A heterogeneous storage grid enabled by grid service , 2007, OPSR.

[13]  ShenkerScott,et al.  Search and replication in unstructured peer-to-peer networks , 2002 .