Volume : III, Issue : IX, September - 2014

Metrics for Measurement of Service Replicas in Grid Environment

P. Sunilgavaskar, Ch. D. V. Subbarao

Abstract :

The important requirement in data grid environment is replicas utilization. In general replicas are used as proactive (i.e. failure considered before scheduling of a job) and post active handles the job failure after it has occurred. We used centralized and local replicas within agents. In this paper, we address the probabilistic models of Fault tolerance, communication, efficiency and reliability of simple job submission. We have shown an autonomous, self adaptive agents model to model replicas in distributed environment for long execution of jobs. The applicability of the proposed agent’s architecture in data grid is able to setup some parameters of replicas such as centralized and local replicas search rate with mean time failure and number of objects used in it. The search values of centralized and local replicas are measured when computing happens. The Metrics used for search value are number of replicas, replication objects with different states and processing capability. The metrics we considered are to reduce complexity in search rate of replicas due to their huge computing and storage capabilities. On the basis of centralized and local replicas search value we predict the unreliable agents computing model as result of that we can increase computing reliability or we can avoid unreliable model into computing. Hence we showed an autonomous, self adaptive agent model in order to reduce service failures of replicas in large scale distributed environment.

Keywords :

Article: Download PDF   DOI : 10.36106/ijsr  

Cite This Article:

P.SunilGavaskar, Ch.D.V.Subbarao Metrics for Measurement of Service Replicas in Grid Environment International Journal of Scientific Research, Vol : 3, Issue : 9 September 2014

Number of Downloads : 655

References :