Modelling resource estimation and scheduling for processing in cloud computing

Cloud computing is nowadays replacing the office based computing. As cloud computing becomes more widespread, the allocation of network and computing resources will be an issue. While data center energy use has received much attention recently, there has been less attention paid to the resource (CPU processing time and bandwidth) estimation in the cloud networks, which helps to connecting users to the cloud. This paper presents a formal analysis of modeling resource (CPU time) estimation in cloud computing, considering private cloud. The estimation model is based on Linear Least Square Regression (LLSR) method. The performance is measured for different applications in a cloud environment, which is setup and tested for resource estimation of user´s frequent applications to be processed in the cloud. The estimation times given by LLSR are compared with other allocation algorithms like WFQ and SWFQ. Though the performances of WFQ and SWFQ are initially matching with the actual CPU time, they are not giving better estimates for the later applications. LLSR gives better estimates which match almost with the actual allocation for all applications.