Hello, Took over a small NetApp Env. We have two FAS3240's running in 7-mode. We are running version 8.2.3. We have about 6 Shelves of disks, a mixture of 15k, 10k, and 7.2k SAS.
We are strictly using it for our Vmware 5.5 Environment running about 100 Server VM's at the moment.
We are not using Snapshots or any other technologies
We have not had any performanace complaints as of yet. So everything is running smoothly. But one of the concerns is Space. The cost of adding more shelves is becoming a factor so we are in the process of rediesigning things around to squeeze more space out of system.
To do this we are redesigning our Volumes and LUNs so that we can start Overcommitting on space. We are doing this by Thin provisioning our Volume and Luns, and then running Dedupe on the Volumes.
Currently ive only been doing this on one of our 7.2k Shelves. Seems to be working fine.
I know by adding Thin Provisioning and Dedeupe we are adding overhead and will take a small performance hit by doing this.
I guess my question is, is there a rule of thumb to how many Volumes can be Thin Provisioned and Deduped before performance becomes an issue? I realize thats a general question as some questions will be, what are you running on your VM's and such. We have a mixture of small database servers, Web Servers, and allot of File Servers. Most of the Database servers are on the 15k and 10k Storage.
Another question I guess would be, how do i monitor performance with Netapp to try and prevent performance before it gets to be a problem. I know that OnCommand System Manager has realtime performance but that only shows things in realtime. I have something called Netapp Oncommand Console, is this what i should be using and what are some common things to look at.
Im going to start looking more into this but thought ide ask the community first.
Thanks for any input on this
