Enabling and optimizing pilot jobs using XEN virtual machines for hpc grid applications

The primary motivation for uptake of virtualization have been resource isolation, capacity management and resource customization: isolation and capacity management allow providers to isolate users from the site and control their resources usage while customization allows end-users to easily project the required environment onto a variety of sites. Various approaches have been taken to integrate virtualization with Grid technologies. In this paper, we propose an approach that combines virtualization on the existing software infrastructure such as Pilot Jobs with minimum change on the part of resource providers. We also present a standard API to enable a wider set of applications including Batch systems to deploy virtual machines on-demand as isolated job sandboxes. To illustrate the usefulness of this approach, we also evaluate the impact of Xen virtualization on memory and compute intensive tasks, and present our results that how memory and scheduling parameters could be tweaked to optimize job performance.