ONTAP Discussions

VMWare volume, 16% space savings, should it be higher?

HendersonD
17,877 Views

I have a 2TB volume on a FAS3020 running OnTap 7.3.2. This volume is shared via NFS and is added as storage for my ESXi 4.1 hosts. I have 29VMs stored on this NFS share, a mixture of Win2003 and Win2008 servers. I have had dedup enabled on this volume for several months. At the command line of the filer I ran:

df -s /vol/VMWare

It says there is a 16% space savings. This seems low. I thought I had read other accounts, as well as Netapp marketing material, that claims space savings for VMWare in the 40-50% range. Any ideas?

I do take snapshots on this volume using SMVI but as I mentioned I have had dedup turned on for months and only keep a few weeks worth of snapshots.

1 ACCEPTED SOLUTION

datamanaged
18,253 Views

Hi Again Hendersond,

I'm not sure if my math is 100% correct here, but its not far from it.

Total volume usage - Snapshot Usage = Live Data

Dedupe Savings / (Live Data + Dedupe Savings) = Dedupe ratio excluding snapshots

For you:

1068 - 773 = 295

244 / (295 + 244) = 45.2% saved

The 19% you're seeing is based off of the 244GB of savings on 539GB of live data with an overhead of the snapshots 773GB (and for some reason it treats the snapshots as undeduped, when in reality they probably *are* deduped). If we factor snapshots into the above equation: 244 / (295+244+773) we get 19% after rounding.

We actually just noticed this with our DFM provisioning manager, whereas the CLI shows the dedupe saving including snapshots as just used storage, DFM seems to exclude snapshots from its computations. Our VSphere volume is ~42% saved in the CLI in DFM its abou 52% ( we have a very short backup schedule)

If you really want to tell, delete your snapshots(prolly not a good idea though). Alternatively, just try deleting a number of snapshots off the tail end (such as all of november if you don't need them). This might give you an idea of how the dedupe % will fluctuate.

Long story short, its a bit of bad math in df -s in combination with your snapshots.

I hope we've answered the conundrum at this point. If you still have problems our questions, let us know.


Best Regards,

Adam S.

View solution in original post

48 REPLIES 48

HendersonD
5,398 Views

Out of my 29VMs, 13 run Server 2003 and 16 run Server 2008 R2. The ones that run 2008 R2 where built from a template made from stratch. A few of the Server 2003 VMs were brought over via a P2V but they were all aligned using Netapp's MBR Align utility. Just to check, over a month ago I brought up an Ubuntu VM to run MBR Align since I am using ESXi which will not run MBR Align in the console (no console!). I checked and all VMs are aligned properly. I may fire up this VM again just to double check.

I have never tried any type of space reclamation tool, is there any one you can suggest?

Dave

HendersonD
5,398 Views

I did find out one interesting thing. I thin provision all of my disks used in my VMs. This uses up space on my storage as needed. If a bunch of files are added to a VM, more space is consumed on the Netapp volume that holds the VM. If these files are later deleted, this space is not reclaimed. To reclaim this space you can use sdelete from Microsoft's sys internals suite. Here is the procedure:

http://www.virtualizationteam.com/virtualization-vmware/vsphere-virtualization-vmware/vmware-esx-4-reclaiming-thin-provisioned-disk-unused-space.html

I did take several of my VMs and reclaimed some space. Now, I have no idea if this will help my dedup savings.

Thanks for the suggestion

keitha
5,398 Views

That can certainly help, all those new zeros dedupe very well, sometimes those ghost files can really mess things up. Keep us posted, once those snaps roll off you should be seeing a much higher number....

Keith

radek_kubka
5,398 Views

Hi,

If a bunch of files are added to a VM, more space is consumed on the Netapp volume that holds the VM. If these files are later deleted, this space is not reclaimed

There is a feature in SnapDrive called Space Reclamation a.k.a. hole punching which does exactly what you are looking for. But I would challenge your expectations it may increase your dedupe ratio:

- hole punching releases zeroed blocks to the volume, so the space is saved, indeed, yet these zeroed blocks will be marked as free in ONTAP, hence won't be a subject for de-duplication if you know what I mean.

Regards,
Radek

HendersonD
5,397 Views

I created another volume on the Netapp to hold application data. I made an NFS share from this volume and then added it to my 4 ESXi servers. I added a drive to 6 of my VMs with the data located on this datastore. These 6 VMs had a significant amount of application data that was located on the same datastore as the operating system (C: drive) but is now located on a separate datatore. In total I moved 110GB of application data off the datastore that holds the boot drives for my VMs.

I also did some cleanup work. Several of my VMs had installers left behind on the C: drive. I deleted these installers, some of which were as large as 4GB

Attached is a spreadsheet showing the space chewed up by each of my VMs. After some more analysis I realized that Windows Server 2008 R2 on average chews up about 10GB of space while WinSever 2003 chews up about 4GB. I really only have 3 VMs left where this is a significant amount of applicaition data on the C: drive, these are highlighted in yellow. Once these three servers are migrated to WinServer 2008 R2 I can separate application data from the OS, this could take several months.

Even though the average amount of space that WinServer 2008 R2 takes up is 10GB, this can vary. If you look at my spreadsheet you will see servers like Exchange and SQL where the OS takes up considerably more space. Much of this is used by Exchange and SQL itself. This IS NOT my databases and log files, but just doing the install of Exchange and SQL onto the C: takes up a lot of space.

The bottom line is the OS is using up about 58% of the space on the VMWare volume while applications and application data about 42%. Even when I get the three VMs highlighted in yellow split out, the best I can achieve will be an even split. Even this is really just an estimate, the bulk of my application data has already been pulled off the VMWare volume. In two weeks once my snapshots age off I will be able to see the affect of this work. My hope is my dedup savings jump from 32% to something above 50%.

sarahj241
5,396 Views

Yeah, I also had a question like this, so thanks, for all the comment help!  These guys also have some good info.

HendersonD
3,325 Views

As I reported in my previous post, I moved a good bit of application data off the C: drive for many of my servers and onto the E: drive which is on a separate datastore. After waiting two weeks for all of my snapshots to age off, my space savings now sits at 50%

I still have 4 or 5 servers that have a significant amount of application data on the C: drive. These will take me several months to cleanup, in a few instances I need to wait until summer to do this work. This should drive the space savings even higher. Thanks to Radek, Evilensky, datamanaged, and keitha for helping out. After struggling with this for 3 months the bottom line to maximimize deduplication space savings:

- Move pagefiles to a separate disk on a separate datastore. Make the disk independent-persistent so snapshots are never taken of these pagefiles

- Move as much application data off the boot (C:) drive and onto a separate drive that is on a separate datastore

Public