The virtual environment based on high performance computing

Abstract

newline Grid computing is an interconnected network of computing resources that works together for the completion of a task instead of relying on a single computer. The grid is created using virtual machines instead of physical machines. The use of a virtual machine grid reduces the hardware cost, reduces the execution time of algorithms, and reduces the communication delay between the nodes inside the grid environment. Oracle VirtualBox and VMware Workstation (type 2 hypervisors) are used for constructing the grid structure in a virtual machine environment. While executing algorithms inside normal grid, high execution time is examined. This high execution time causes performance degradation, so in order to overcome this, virtual machine grid environments are created. In the proposed work the performance can be improved by reducing the execution time of the subtask inside grid, reducing power consumption by running different virtual machines inside a single server and reducing the hardware cost by creating different virtual machines inside a single physical server. For evaluating the performance of the virtual machine grid environments, the firefly algorithm, Ramanujan number, and square of numbers are run in the virtual machine grid and ordinary physical machine grid separately. The execution time of algorithms, the communication delay between the grid nodes, and the hardware cost can be reduced by using a virtual machine grid. The performance of the virtual machine grid can be again improved by using the type 1 hypervisor (KVM) and the addition of more powerful cloud nodes from the open nebula. The execution time of the algorithms is again reduced, which is already executed in type 1 hypervisors iv like oracle virtual box and VMware workstation. To harness the power of cloud, grid, and cluster computing, jungle computing has been introduced.