Essential Guide

Best practices, tips and tools for VMware virtual recovery and backup

A comprehensive collection of articles, videos and more, hand-picked by our editors

VMware snapshot size and other causes for slow snapshots

Snapshots offer multiple restore points for VMs, but add complication. Poor performance can stem from the VMware snapshot size and other root causes.

VMware snapshots occasionally become slow, which drags down the virtual machine's performance. You can prevent

or fix slow VMware snapshots. If that doesn't work, there are ways to avoid snapshots altogether.

To resolve slow snapshots, you need to find out what's holding up your snapshot procedure. Is the virtual machine (VM) busy? Is its storage overtaxed? Could something else in the infrastructure be causing a snapshot problem?

Understanding VMware snapshots will help you know where to look for root causes. In VMware virtual environments, snapshots are metadata copies of the entire state of the virtual machine. The snapshot doesn't contain much data, mainly creating pointers to the original VM. When the VM state changes, the snapshot grows in size with information on the changed data blocks.

Also consider why you are creating snapshots. Snapshots are used for a reliable backup procedure, as well as restoring a virtual machine to a specific past state. These are completely different cases when troubleshooting slow snapshots. In some instances, snapshots might be the wrong tool for the job.

Become a snapshot expert

Automate snapshot consolidation

Troubleshoot problems with VMware snapshots

Take our snapshots quiz

The longer a snapshot is around, the bigger it will get, and bigger means slower operations. To keep VMware snapshot size within recommended parameters, delete snapshots no later than 72 hours after creation. This is probably the top resolution for slow VMware snapshots: Delete them before they grow so big! Especially in test environments, admins will stack multiple snapshots so they can easily revert to a VM's previous state. Consider alternatives such as cloning a virtual machine once in a while, if you want to keep around a much earlier state of the VM.

The best approach for clean and reliable snapshots is to stop, pause or shut down a VM before taking its snapshot. This is the only method that ensures that no other processes are accessing the virtual machine. Do this whenever possible, although it is unacceptable for some workloads to take the entire VM offline.

Slow snapshots can also occur when a VM has insufficient resources. Check that the resources allocated to the virtual machine can handle a normal workload with the addition of snapshot processes. Analyze the virtual machine's performance to see if it has sufficient CPU cycles, memory and I/O to handle the snapshot procedure. Simple changes in VM configuration may also help. For instance, if the operating system is on a different disk image than the virtual machine's data, the snapshot process will finish much faster on the OS portion, allowing it to resume normal operation while the snapshot finishes.

If these VMware snapshot best practices don't improve VM performance, consider approaching your task with a different tactic. If you're using virtual machine snapshots to make the backup procedure easier, for instance, you could switch to an agent-based backup. This alternative to snapshots and agentless backup has its own drawbacks, with more resource consumption than what you're used to with agentless backup.

This was first published in May 2013

Essential Guide

Best practices, tips and tools for VMware virtual recovery and backup

0 comments

Oldest 

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to:

SearchServerVirtualization

SearchVirtualDesktop

SearchDataCenter

SearchCloudComputing

Close