Abstract: The increasing popularity of Cloud Computing as an attractive alternative to classic information processing systems has increased the importance of its correct and continuous operation even in the presence of faulty components. But fault tolerance is still a crucial issue in SOC. By utilizing VM technology, the performance and fault isolation of cloud resources can be provisioned on demand in a fine grained, multiplexed manner. The design of the DOPS resource allocation method, utilizes the proportional share model. The key idea is to redistribute available resources among running tasks dynamically, where these tasks could use up the maximum capacity of each resource in node. To locate for qualified nodes in the SOC environment, a fully decentralized range query protocol called P-G CAN has been tailored for DOPS. Pointer Gossiping is responsible for making sure every node in the system eventually knows important information about every other node’s state including those that are unreachable. The sensitivity analysis conducted on the model, violates certain assumptions that creates more impact on the dynamic optimal resource allocation of multiple attributes due to faulty components. Hence an analysis will be made in a VM environment, aiming to find a node already present in the network and identify a zone that can be split. Thereby, updating the routing tables of nodes neighbouring the newly split zone. During this step, faulty nodes will be identified and re-routing will be done either by merging zones or managing with the neighbour node, all performed dynamically by the SOC system.

Keywords: Self –Organizing Cloud (SOC), Virtual Machine (VM), Dynamic Optimal Proportional Share (DOPS), Pointer-Gossiping Content Addressable Network (PG-CAN), Sun Grid Engine Cluster (SGE).