Cloud Resource Allocation Schemes Using Drip Irrigation Approach

Managing resources in cloud introduce the challenge of on-demand and optimistic resource allocation in response to service requests. Typically cloud computing offers an elastic infrastructure that resource provider agent can use to obtain resources (e.g., hardware, software resources) that meet the demand. Resource provider agents are charged for the amount of resources allocated in the cloud. In this respect, an issue is to decide on the right amount of resources reserved in the cloud, and its reservation time so that the financial cost on the resource provider agent is minimized. This paper proposes a simple and easy method to implement algorithms for efficient resource allocation schemes that exploit optimistic rates while ensuring that adequate resources are reserved in the cloud for the clients. Based on the prediction of demand for the request, proposed algorithms are designed by simulating drip irrigation approach in order to mitigate the wrong resource allocation decisions. The results of experimental evaluations show that the proposed algorithms significantly reduce the financial cost of resource allocations in the cloud as compared to other conventional schemes.