Virtual Machine Placement With Automatic Deployment Error Recovery

A method comprising:

• receiving a request to select at least one candidate datastore for a plurality of virtual machines (VMs), said request identifying at least one host and resource requirements associated with the plurality of VMs;
• selecting the candidate datastore from a plurality of datastores accessible to the host based on a comparison of the resource requirements and an assessment of resources associated with the plurality of datastores;
• deploying at least one of the plurality of VMs on the host and the selected candidate datastore;
• updating, during said deploying, the assessment of the resources;
• and
• selecting another candidate datastore accessible to the host based on a comparison of the resource requirements and the updated assessment of the resources.

One or more computer-readable storage media including computer-executable instructions that, when executed, cause at least one processor to place virtual machines while implementing error recovery by:

• receiving a request to select at least one candidate datastore for a plurality of virtual machines (VMs), said request identifying at least one host and resource requirements associated with the plurality of VMs;
• selecting the candidate datastore from a plurality of datastores accessible to the host based on a comparison of the resource requirements and an assessment of resources associated with the plurality of datastores;
• deploying at least one of the plurality of VMs on the host and the selected candidate datastore;
• maintaining, during said deploying, a status log for each of the plurality of VMs;
• detecting deployment failure during said deploying;
• and
• resolving the detected deployment failure based at least on the maintained status log.