IPOS for NetAPP or IBM n series SAN

Does anybody have a nice link to the IOPS for an n-series 6040 (IBM's name) or the equivalent for NetApp?

We're looking at a few SANs and that is one of them, but they seem to be hiding the performance numbers.
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.

Duncan MeyersCommented:
The equivalent NetApp array is a FAS3140. The system can support up to 480 drives, so theoretical maximum IOPS with 15K Fibre Channel drives is 96,000 IOPS. Fully populated with 10K FC drives, and about 34,000 IOPS for SATA drives (again, fully populated with 480 drives), maximum theoretical IOPS is in the order of 57,000. The real numbers will typically be less than that as you lose some drives for system volumes, parity, snapshot reserve and so on. Also, typically an array is configured with different drive types; a mix of 15K FC, 10K FC and SATA drives is normal, and that will affect the total IOPS that the system can deliver.

It's a pretty serious box.

You'll find that the enterprise SAN vendors no longer quote performance figures - I think they all found themselves in sticky situations having ot justify their performance figures, and the reality is that to hit the upper end of the quoted performance figures requires an array that is configured to provide maximum performance - nothing else. One of the second tier SAN vendors (since absorbed into one of the big PC vendors) used to quote "Up to 60,000 IOPS!" which was a complete fiction - not in their wildest imaginings could 16 drives produce that sort of performance. I note with interest that they no longer make the claim.

As the number of drive spindles is the main determinant of performance, you can calculate the potential performance from the number of drives and IOPS of each drive type. Use 200 IOPS for 15K FC, 120 IOPS for 10K FC and 70 IOPS for 7200rpm SATA drives.

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
cajxAuthor Commented:
Thanks, that is exactly the kind of information I was looking for!
cajxAuthor Commented:
This 2nd tier vendor wouldn't happen to start with the letter L and end in the letter D would it? If so, those are the guys we were looking at also.
Determine the Perfect Price for Your IT Services

Do you wonder if your IT business is truly profitable or if you should raise your prices? Learn how to calculate your overhead burden with our free interactive tool and use it to determine the right price for your IT services. Download your free eBook now!

Duncan MeyersCommented:
Four letter word  - starts with D, ends with L. Rhymes with Swell. The storage vendor is from an acquisition they made about 18 months ago.

The big thing that NetApp (or N series IBM) has that no-one else has is brilliant snapshots - and the ability to use those snapshots for backups with SnapVault. It's seriously cool stuff. NetApp also have some cool stuff they do with VMware with SnapManager for Virtual Infrastructure. Ultimately, with the tier one vendors, the storage is all much of a muchness. It's the software bolted on top that makes the difference between the vendors. NetApp have brilliant snapshots, EMC has brilliant replication (RecoverPoint), path balancing software and probably the best maintenance service in the industry, HDS has cache partitioning and the best storage virtualisation platform (USP-V), and so on. Personally, I'm a big fan of EMC CLARiiON and NetApp.

And thanks for the points! Glad I could help.

Duncan MeyersCommented:
From the sounds of it, you won't  - but beware of falling into the trap of sizing the array for amount of storage space you need without accounting for performance. Unless you understand the performance requiorements of your environment, it's too easy to say: "we need 20TB, so all we need is 22 x 1TB drives" - yet the underlying performance needs might (and probably will) need 3 or 4 times the number of drives. It's not always easy to explain the concept to management, either...
cajxAuthor Commented:
OK, we weren't look at the one that rhymed with swell, but I appreciate the heads up!

I have been reading that more small drives (spindle count) is a biggie. For our environment, we are not huge. We have about 250 exchange users... but only about 120 are at the office that will be going SAN/VMware. We have about 100 MS SQL users and 40 Citrix users. Those boxes are the ones we are the most concerned about for performance. Number of users is of course one thing, but what they do is another. We have had SQL performance issues with our current ten count of local RAID 5 drives. Some of it could be software, but some of it could be accessing time.

Anyway, I had a smart fellow in the vmware forums also point to NetApp being better on latency than Lefthand (our alternate). But he made the point that they should allow us to test drive something to see if it performs well enough for our environment. The LeftHand mirroring is pretty sweet for the cost, so we haven't discounted them just yet. But performance is a big one!
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

From novice to tech pro — start learning today.