Abstrakt

Efficient Cloud Computing Load Balancing Using Cloud Partitioning and Game Theory in Public Cloud

P.Rahul, A.Senthil Kumar, Boney Cherian

Cloud computing could be an idea that has several computers interconnected through a true time network like web. Cloud computing is economical and ascendable however to keep up the soundness of process many roles within the cloud computing could be a terribly tough drawback. The work arrival pattern cannot be foretold and also the capacities of every node within the cloud disagree. Thus for levelling the usage of web and connected resources has enhanced wide. As a result of this there's tremendous increase in employment. Therefore there's uneven distribution of this employment which ends in server overloading and should crash. In such the load, it's crucial to manage workloads to boost system performance and maintain stability. The load on each cloud is variable and addicted to numerous factors. Smart load levelling makes cloud computing additional economical and conjointly improves user satisfaction. The load levelling model given during this work is aimed toward the general public cloud that has various nodes with distributed computing resources in many alternative geographic locations. Thus, this model divides the general public cloud into many cloud partitions. Once the atmosphere is extremely massive and complicated, these divisions alter the load levelling. The cloud encompasses a main controller that chooses the acceptable partitions for incoming jobs whereas the balancer for every cloud partition chooses the most effective load levelling strategy.