Scheduling to Minimize Context Switches for Reduced Power Consumption and Delay in the Cloud

Abstract

Cloud computing is an emerging technology to increase the efficient utilization of huge collection of high-end resources at low cost. One of the major problems associated with cloud environment is that these high-end systems need huge power consumption. Several methods are proposed and available for reducing the power consumption in datacenters. But most of these methods do not meet the Quality of Service (QoS) requirements such as minimum delay during packet transfer with reduced power consumption. Data center contains many physical machines each of which can be virtualized into several virtual machines (VM) according to the needs. Virtualization technique virtualizes the physical resources for satisfying the user requests. Proposed system considers power consumption and delay during packet transfer in a single physical machine. Two schedulers work concurrently to schedule both Virtual Machines and processes. A new scheduling algorithm is implemented for both VM Scheduling and process scheduling based on the buffer size of the port. Proposed algorithm tries to reduce the waiting time as well as the number of context switches. By reducing the waiting time and context switches, system can reduce delay and power consumption during the packet transfer between VMs.