Fault Tolerant Decentralized Scheduling Algorithm for P2P Grid

Abstract Peer to peer (P2P) grid systems consist of nodes present in multiple organizations and spread across different physical locations. These resources join and leave P2P grid depending upon their determination. Therefore, availability of grid resource cannot be guaranteed at all times. Moreover, failure in network infrastructure, hardware and software of resource can force it to leave grid without any early warnings. Many authors have proposed decentralized scheduling algorithms for P2P grid. In this paper, we have proposed a fault tolerant decentralized scheduling algorithm for P2P grid, which reschedules jobs of grid resource when node failure happens. This mechanism handles node failure in fully decentralized manner. Key feature of this fault tolerant approach is that, it reschedules jobs depending upon communication and computation cost associated with jobs.

[1]  Miron Livny,et al.  Condor-a hunter of idle workstations , 1988, [1988] Proceedings. The 8th International Conference on Distributed.

[2]  Guillaume Pierre,et al.  On the Feasibility of Decentralized Grid Scheduling , 2008, 2008 Second IEEE International Conference on Self-Adaptive and Self-Organizing Systems Workshops.

[3]  Jason Maassen,et al.  Zorilla: a peer‐to‐peer middleware for real‐world distributed systems , 2011, Concurr. Comput. Pract. Exp..

[4]  Francisco Vilar Brasileiro,et al.  Faults in grids: why are they so bad and what can be done about it? , 2003, Proceedings. First Latin American Web Congress.

[5]  Nitin,et al.  Decentralized Computation and Communication Intensive Task Scheduling Algorithm for P2P Grid , 2012, 2012 UKSim 14th International Conference on Computer Modelling and Simulation.

[6]  A.A. Azab,et al.  An adaptive decentralized scheduling mechanism for peer-to-peer Desktop Grids , 2008, 2008 International Conference on Computer Engineering & Systems.

[7]  Manoj Kumar Mishra,et al.  An analysis of various job scheduling strategies in grid computing , 2010, 2010 2nd International Conference on Signal Processing Systems.

[8]  Maarten van Steen,et al.  CYCLON: Inexpensive Membership Management for Unstructured P2P Overlays , 2005, Journal of Network and Systems Management.

[9]  Jie Xu,et al.  Fault Tolerance within a Grid Environment , 2003 .

[10]  Franco Zambonelli,et al.  Co-Fields: A Physically Inspired Approach to Motion Coordination , 2004, IEEE Pervasive Comput..

[11]  Yi Yang,et al.  Computing Field Scheduling: A Fully Decentralized Scheduling Approach for Grid Computing , 2011, 2011 Sixth Annual Chinagrid Conference.

[12]  Guillaume Pierre,et al.  Decentralized As-Soon-As-Possible Grid Scheduling: A Feasibility Study , 2010, 2010 Proceedings of 19th International Conference on Computer Communications and Networks.

[13]  Larry Carter,et al.  Autonomous protocols for bandwidth-centric scheduling of independent-task applications , 2003, Proceedings International Parallel and Distributed Processing Symposium.

[14]  Mario Lauria,et al.  Self-Organizing Scheduling on the Organic Grid , 2006, Int. J. High Perform. Comput. Appl..

[15]  Soonwook Hwang,et al.  A Flexible Framework for Fault Tolerance in the Grid , 2003, Journal of Grid Computing.

[16]  Ian T. Foster,et al.  Application experiences with the Globus toolkit , 1998, Proceedings. The Seventh International Symposium on High Performance Distributed Computing (Cat. No.98TB100244).

[17]  Dick H. J. Epema,et al.  KOALA: a co‐allocating grid scheduler , 2008, Concurr. Comput. Pract. Exp..