Systems and methods for fault tolerant batch processing in a virtual environment

摘要

A system for fault tolerant batch processing in a virtual environment is configured to perform batch job execution, the system includes computing devices configured as a virtualized grid cluster by means of a virtualization platform, the cluster includes a centralized storage repository, a grid manager deployed on an instantiated virtual machine and a message bus whereby data and messages are exchanged between the grid manager and one or more grid nodes. The grid manager is configured to manage one or more incoming job requests, queue one or more of the received job requests in a job execution queue and monitor one or more virtual grid nodes.

主权项

1. A method executed by one or more computing devices for batch processing in a virtual environment the method comprising:
identifying by at least one of the one or more computing devices, a batch job for execution, the batch job including a specification template specifying the required execution environment for the batch job and a batch application; determining, by at least one of the one or more computing devices, one or more virtual machine images based at least in part on the required execution environment; creating, by at least one of the one or more computing devices, one or more batch job-specific virtual machine images based at least in part on the one or more virtual machine images and the batch application; and generating, by at least one of the one or more computing devices, a batch job definition corresponding to the batch job, the batch job definition including an allocation of predetermined computing resources required for completion of the batch job; and queueing, by at least one of the one or more computing devices, the batch job for execution on a job queue.