A Replica Distribution Based Fault Tolerance Management For Cloud Computing Ajitabh Mahalkari, Prof. Ritu Tondon Department of Computer Science & Engineering TRUBA college of Engineering and Technology, Dewas bypass, Indore (M.P) -452001 India Abstract: Cloud computing now a day's become most popular and reliable computing technique for organizations and individuals. In the cloud environments, data availability and backup replication are critical and complex issues in the an efficient fault tolerance policy is the major. Fault tolerance policy is the strategy in action when a fault occurs in the system. Taking backups is the one of the most usual solution of keeping the data safe out of these faulty conditions, But sometimes traditional schemes and backup servers might get into bottleneck conditions due to their replica management schemes. CSP has to adopt a fault tolerance policy to ensure continues delivery of services. There are two main categories policies that are proactive and reactive fault tolerance, both have different advantages in different situations. Our research introduces a new approach as a fault tolerance policy, which is having characteristics of both the above mentioned policies. In our approach replica distribution and retrieval is the functional areas. Index Terms: Cloud Computing, Fault Tolerance, CSP, Replica, I. INTRODUCTION Cloud computing is modern computing model based on on-demand services. For organizations to be competitive in frequently changing market and highly interconnected global economy, they need to be capable enough to be able to respond rapidly to the changing market. A cloud, a next level computing model, provides amazingly scalable and flexible computing that is available on demand. Cloud Computing facilitates self-service requesting through a user friendly interface. Cloud Computing promises real costs effective and agility to organizations. Cloud computing serves the demands of a number of individuals and organizations. Demands may be higher end services, data or any other computing resources. Consumer needs are varied and challenging to full fill. Customers do not likely to bear service downtime and data unavailability without keeping technological challenges in mind. Customer needs include high performance and data availability. Fault tolerance is a major concern to the service provider to assure availability of data and continuity of services. Failures should be anticipated and handled proactively or reactively according to the service and the level of agreement with the consumer. Fault tolerance techniques are used to predict these failures and take appropriate precautions, or remedial actions before failures can actually occur. Effective and efficient fault tolerant techniques are need of cloud computing. We are trying to introduce a combination of proactive and reactive fault tolerance policy in this paper. Data loss occurs for different reasons reduce availability of the data and retrieval causes degradation in reliability and authentication over the system. To deals with such issues cloud computing needs an effective storage and retrieval policy requires dedicated fault tolerant mechanism which makes the system robust and safe. These legacy fault models distribute the loads among the various nodes and processing routine execution time reduces dynamically. Every methodology would work according to their reliability of operations. Successful data backup and retrieval of data after faults eventually increases the system reliability and decreases after crash or unavailability. The reliance, values changes after each computing cycles and measures the behavior of the system. The process of assigning the reliability values for system is a time based approach in which the maximum or minimum reliability decides the kind of data backups required. Thus it can be considered as a key decision making approach for selecting the backup policy. Formally, it says for minimum reliability values the pre-emptive methods must use and the protective mechanisms are used for maximum values of reliability. To satisfy the reliable operations with fault tolerant and robust system, backups are required. These backups are provided on the basis of their backup condition which gives the ratio of backup needed. This ratio is measured in terms of the replica policies. Here the replica is the copy of data which is continuously exchanged between the different Ajitabh Mahalkari et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 5 (5) , 2014, 6880-6887 www.ijcsit.com 6880