• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 314
  • Last Modified:

Netapp VMDK eager zero Best Practices

I am looking for clarification on the best practices in using the Netapp ontapp 8.2.2 with VM~thin provision.  My understanding is the best practice is to make sure the flex volume is thin provisioned and that the Lun is also thin Provision.  However what I don’t understand is what the best practice is at the VMDK level.  My understanding was that you would set the hard disc as Thick  eager Zeroed  This will write Zeroed data to the hard disc but the Netapp through VAAI connection would understand what was happening at the ESXI level and in turn not allocate the space at the Agg level.  However my testing here is showing that when we thick provision eager Zero the space also gets allocated at the Netapp aggregate level.  I have confirmed as expected that thin provisioned and Thick Lazy  at the VMDK does not allocate the space at the netapp.  .  What I’m trying to understand is whether I have configured the VAAI connection incorrectly or whether this is the expected behavior.  This is being used as a iscsi not NFS
Your thoughts are greatly appreciated.
0
James-Sillett
Asked:
James-Sillett
1 Solution
 
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
Thick eager Zeroed virtual disks, have been proved to yield performance figures of 10-15% better than lazy zeroed.

If you are concerned with top performance, and clustered disks, or FT virtual machines, I would opt for Thick eager Zeroed virtual disks.

All our clients with NetApp Filers use  Thick eager Zeroed virtual disks, however, there is a delay in the initial creation of the virtual machine, if creating from scratch, however, we use templates for deployment of virtual machines, so the initial creation of the golden master, is the only initial delay.

Have you added the VAAI plugin ?

This web article, shows some tests

http://www.virtuallanger.com/2011/12/06/vaai-is-this-thing-on/
0
 
gheistCommented:
With iSCSI it supports DISCARD/TRIM, and netapp on the backend does its own thin provisioning/dedupllication as configured per volume., not per LUN.
So you have the choice between vmware handling thin provisioning, or in case your guest does TRIM/DISCARD (Like Windows 7 or linux 3.something) you can rely on netapp.
Why not stick with default lazy zeroed provisioning? It is fast to create.
0

Featured Post

Free Tool: Site Down Detector

Helpful to verify reports of your own downtime, or to double check a downed website you are trying to access.

One of a set of tools we are providing to everyone as a way of saying thank you for being a part of the community.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now