NetApp - What type of disks for additional shelf?



I have a NetApp FAS 2040 version 7.3.3 filled with Twelve 3.5 Inche drives at 15K and 600GB. I need another shelf with more disks as i'm running out of space. I really don't need all the IOPS. Our NetApp sales guy and NetApp support are telling me to go with 15K drives on the new shelf which is the DS4243 as shown.

Which of the two models do you think i should go with from the link below (DS2246 or DS4243) and why?

http://www.netapp.com/us/products/storage-systems/disk-shelves-and-storage-media/disk-shelves-comparison.html
LVL 1
First LastAsked:
Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Andrej PirmanCommented:
Are you aware that shelf unit is 3-times the price of base unit you already have?
I mean, does price play a significant role?
0

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
First LastAuthor Commented:
I was told I have to purchase at minimum 12 disks with my shelf. This is a lot cheaper than going with 24 disks. However I don't really need the IOPS of 15K, but need the space.  I average 300-600 IOPS on my current setup. From testing I know the system can handle roughly 3000+ IOPS.

The DS4243 with 12 disks at 15K is roughly the cost of the FAS2040.

Price is a huge factor.
0
Paul SolovyovskySenior IT AdvisorCommented:
If you don't need the IOPS don't go for the 15k spindles.  I typically have a SATA shelf for CIFS shares and low IOPS VMs

Bigger question is why you need to added storage?

Depending you on your configuration and if you haven't done any space reclamation on your LUNs (think snapmanager for exchange, sql) you can get back a lot of storage.  For most customers that haven't performed the maintenance required we have been able to get back 20% to 50% on from current storage..one customer we were able to save as much storage as a DS4243 shelf.

I would do a basic assessment and use that as my baseline.
0
Ultimate Tool Kit for Technology Solution Provider

Broken down into practical pointers and step-by-step instructions, the IT Service Excellence Tool Kit delivers expert advice for technology solution providers. Get your free copy now.

First LastAuthor Commented:
I'm using iSCSI and was told Snapmanager only works on NFS. I've heard this through NetApp support and our reseller. Did something change they aren't aware of? I know they aren't always right and do appreciate your input.

I did a manual space reclamation. I put each of the larger VM's into their own lun. Then delete the old LUN. From there there system ran an automate cleanup. I'm sitting at roughly 75% usage after reclamation instead of the 99%. From what I read one shouldn't be above 75% usage and i see why. When we bought the system we were using roughly 40% of the storage, but as in all things when resources are there they get used. We brough a lot of items in house that were hosted outside and implemented new VM systems that were pending. In planning I expected the current setup to last two years, but we only made it a year.

We did have logging cranked up on all our systems since getting the SAN. That was taking up 10GB per month, but those files are now zipped monthly down to 100MB saving us space.
0
robocatCommented:

You answered the question yourself: if you don't need IOPS, go for the 10K drives.

However, if Netapp support recommends the high end drives, you should find out why. They normally don't recommend an oversized solution unless there's a good reason (or perhaps they don't have all necessary info).

The tip from paulsolov is a good one, that doesn't cost you anything but a little time.

Another, unrelated tip: upgrade from 7.3.3 to 7.3.6. as soon as possible. You really don't want to be running 7.3.3.


0
Paul SolovyovskySenior IT AdvisorCommented:
Are you running Exchange and SQL with Snapmanager with LUNs that sit outside of the virtual machine?  These LUNs (especially the log volumes) can normally be used by snapmanager to reclaim storage.

0
First LastAuthor Commented:
I'm not sure if i'm understanding correctly. I'd say my answer is No. Our LUNS are mapped via VMware to the NetApp 2040. The OS along with all the data for Exchange and SQL are in virtual disks provisioned within the lun. I'm not doing any direct mapping to LUNs provisioned on the NetApp from inside the virtual machines.
0
robocatCommented:

Just make sure you don't keep more volume snapshots than you actually need. Check this with the system manager.

You could also try enabling deduplication to create more space savings.

Did you ask the netapp why they recommend the high end drives ?
0
First LastAuthor Commented:
NetApp didn't give me a good reason why they wanted me to use the 15K drives. They just said it would keep everything uniform so I can add disks to my current volumes.

What I did:

zero out the free space on my VM's with a sysinternals tool
use vCenter Converter to shrink VM's. I only give the VM's 20% free space.
The VMs were already thin provisioned.
Moved larger VMs to their own dedicated LUNs
Delete old luns.
Ran netapp dedup.

I went from 99% storage used down to 47% usage. Deduplication is roughly at 40%. Instead of having two LUNS now I have 8 LUNS. Things are much better!
0
Paul SolovyovskySenior IT AdvisorCommented:
15K is nice for high iops VMs but if you are going to use it for slower storage then you can go with SAS 10K or SATA.  With respect to Netapp it's now how much storage you have it's how much you use it.

You may want to look at NFS for VMWare, grow and shrink volumes on the fly and recovery is easier since you go into snapshot folder if you don't have flexclone and just copy and paste VM back into production datastore, mount without nic and test.  Faster than SMVI and iSCSI although you need flexclone to use single file recovery with VMs on NFS vs iSCSI.

0
Paul SolovyovskySenior IT AdvisorCommented:
Don't forget to use DFM if you have it, it will give you lots of insight.
0
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Storage

From novice to tech pro — start learning today.

Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.