An efficient method for fault tolerance in cloud environment using encryption and classification

Abstract

Cloud computing may be defined as management and provision of resources, software, application and information as services over the cloud which are dynamically scalable. Fault tolerance includes all the techniques necessary for robustness and dependability. The main advantages of using fault tolerance in cloud computing include failure recovery, lower costs and improved standards in performance. Even though the benefits are immeasurable, the element of risk on user applications due to failure remains a major drawback. So our suggested technique utilizes the effective fault tolerance method with the encryption algorithm. To improve the security of the recommended technique, triple-DES encryption algorithm is employed before the data transmission. For the transmission of encrypted data, the implemented method selects the minimum fault tolerance node. So the recommended technique utilizes the effective classification technique. Here, improved support vector machine (ISVM) classifier is used to classify the nodes based on its feature value and the content similarity each node. The proposed ISVM helps in predicting the faults if available, earlier before it occurs. The various parameters considered in our proposed system are accuracy, service reliability and availability. In the proposed method, the accuracy value of the fault tolerance is 79% which is better than in the existing method. The proposed method will be implemented in JAVA with CloudSim.