International Journal of Computer
Trends and Technology

Research Article | Open Access | Download PDF

Volume 4 | Issue 3 | Year 2013 | Article Id. IJCTT-V4I3P105 | DOI : https://doi.org/10.14445/22312803/IJCTT-V4I3P105

Performance and Dependency Analysis of Data in Distributed Systems

Anitha.G.S, Sivagami.V.M

Citation :

Anitha.G.S, Sivagami.V.M, "Performance and Dependency Analysis of Data in Distributed Systems," International Journal of Computer Trends and Technology (IJCTT), vol. 4, no. 3, pp. 219-224, 2013. Crossref, https://doi.org/10.14445/22312803/IJCTT-V4I3P105

Abstract

Atomic Broadcast is important in fault-tolerant distributed computing. It ensures that messages broadcast by different processes are delivered to all destinations in same order. Token circulation is one of the important ordering mechanisms in atomic broadcast. A single proposal is contained in the token. The proposal contains a batch of messages to be ordered and a decision on a batch can be taken at the earliest f (fault) communication steps after the batch is proposed. In the existing system, token based atomic broadcast algorithm relay on the group membership service. This algorithm helps in ordering but it does not tolerate the failure and also a wrong suspicion can lead to two costly membership operations namely addition and removal of process. To solve this problem, token-based atomic broadcast algorithm that uses an unreliable failure detector instead of a group membership service is used .It is efficiently implemented by combining a failure detector and a token-based mechanism. Overhead of a wrong failure suspicion is low when compared to group membership service. The performance of this algorithm is evaluated in both local and wide area networks. The new token-based algorithm provides the better performance of the other algorithms in most small-system settings.

Keywords

Atomic Broadcast, Fault tolerance, Token.

References

[1] Rachid Guerraoui, Ron R. Levy, and Bastian Pochon, “Throughput Optimal Total Order Broadcast for Cluster Environments” ACM Transactions on Computer Systems, Vol. 28, No. 2, Article 5, Publication date: July 2010. 
[2] I. S. Jacobs and C. P. Bean, “Fine particles, thin films and exchange anisotropy,” in Magnetism, vol. III, G. T. Rado and H. Suhl, Eds. New York: Academic, 1963, pp. 271–350.
[3] Parisa Jalili Marandi et.al, “Ring Paxos: A High-Throughput Atomic Broadcast Protocol” in Dependable Systems and Networks, pp. 527-536, 2010K. Elissa, “Title of paper if known,” unpublished. 
[4] Luiz Angelo Steffenel et.al, “Total Order Broadcast on Pervasive System”, ACM Transactions on Distributed systems, SAC’08,2008
[5] Chao-Chin Wu and Ren-Yi Sun, “An integrated security-aware jobscheduling strategy for large-scale computational grids”, Department of Computer Science and Information Engineering, National Changhua University of Education, Changhua City 500, Taiwan. 
[6] P. Urba´n, I. Shnayderman, and A. Schiper, “Comparison of Failure Detectors and Group Membership: Performance Study of two Atomic Broadcast Algorithms,” Proc. Int’l Conf. Dependable Systems and Networks (DSN), pp. 645-654, June 2003.
[7] Marco Primi et.al, “High Performance State-Machine Replication” in  Dependable Systems and Networks - DSN, pp. 454-465, 2011.
[8] Wenbing  Zhaoet.al,“ The Low Latency Fault Tolerance System” Journal:Computing Research Repository - CORR , vol. abs/1004.1, 2010.
[9] J.M. Chang and N. Maxemchuck, “Reliable Broadcast Protocols,”ACM Trans. Computer Systems, vol. 2, no. 3,                pp. 251-273, Aug. 1984.
[10] F.Cristian, S. Mishra, and G. Alvarez, “HighPerformanceAsynchronous Atomic Broadcast,”Distributed System Eng. J.,vol. 4, no. 2, pp. 109-128, June 1997.