A latency reduction mechanism for virtual machine resource allocation in delay sensitive cloud service

Resource provisioning in cloud is highly crucial to confirm the delivery of delay sensitive service through cloud like business transaction etc. Reservation into virtual machine takes place based on cloud resource availability. VM resource would be waste because a large number of virtual machine resources are often provisioned to confirm service response time. There were various techniques to measure service response time but these are for selected cloud infrastructures, analyzing and gathering to collect each dataset individually. Few techniques are perform better in offline analysis but not effective for real-time performance measurement. Now, we are proposed an improve and effective light weight mechanism for real time service latency prediction for optimum virtual machine resource allocation in delay-sensitive services of cloud. Our main aim is to present actual and accurate time delay and cloud resource conditions in short reflect time by our latency prediction mechanism. We are expecting to get 80-90% accurate service latency prediction in cloud resources.