Virtual Machine Integration & Fault Recovery in Crowd-Resourcing Virtual Laboratory

Abstract

Crowd-Resourcing virtual laboratory (CRVL) uses some resources which were shared by the crowd as the contributors to run virtual laboratory exercises. The shared resource is in the form of a virtual machine (VM) that must be automatically integrated into the virtual laboratory system. In this paper, we propose a mechanism of integrating a VM into a CRVL system. It starts with verifying the VM integrity, followed by configuring the VM until the VM is ready to be used to run a Virtual Laboratory exercise. We use WebSocket for communication between the VM and the Virtual Laboratory Server, to be able to monitor the VM Integrity. We also propose a VM fault recovery approach that could automatically divert user connections to other VMs. For the fault recovery purpose, we use SSHFS to mount a remote file system on a file server to store the data of a VM. Our experiments showed that our approaches could work appropriately, and the time for Team Recovery is similar to the time for Team Creation.