When attempting to clone exceptionally large Virtual Machines (e.g., 20TB), administrators often encounter constraints regarding storage space on a single datastore, uncertainty around completion timelines, and questions regarding task management (such as canceling an in-progress clone). This article addresses how to distribute a cloned VM across multiple datastores, estimates the time required for a massive clone and outlines best practices for managing the task safely.
Clone a single Virtual Machine and distribute its virtual disks across multiple datastores. To configure this during the cloning wizard:
Initiate the clone by selecting the Clone to Virtual Machine option.
Proceed through the wizard to the Select Storage page.
At the top of the page, choose the Configure per disk option.
Select individual disks and click Configure. This allows you to assign a specific, separate datastore to each selected disk.
Repeat this process for any remaining disks to distribute the large storage load.
Providing an exact duration for a cloning operation is not possible due to fluctuating environmental behaviors and active loads. However, you can estimate the timeline by understanding your storage limits.
Key factors impacting this duration include:
esxtop command-line utility to monitor how many commands per second (cmds/sec) are currently being executed. This provides real-time insight into the storage load and helps contextualize the transfer speed.It is not recommended to cancel a running clone task. Interrupting a clone operation mid-flight can corrupt or negatively impact the virtual machine files on both the source and the target datastores. Once initiated, you should allow the task to run to its natural completion.