Around the Storage Block Blog
Find out about all things data storage from Around the Storage Block at HP Communities.

Smoking hot midrange SPC-1 results with HP 3PAR StoreServ 7400

CJZ Headshot fixed 150 x 150.jpgBy Calvin Zito, @HPStorageGuy  vexpert 2012 logo.gif

I'm taking a few days off to celebrate my daughter's high school graduation - I'm one proud Dad!  While my level of pride in the latest SPC-1 results don't compare to how I feel about my daughter and all she's accomplished, I wanted to drop by and leave a short blog post about our latest HP 3PAR StoreServ 7400 SPC-1 results.  BTW, she was in a virtual tie for class valedictorian but the tie breaker was credits earned so she didn't get that honor.  Yes, I'm a proud Dad!

 

In the coming days, I hope to do a podcast with our benchmarking expert Paul Haverfield.   I did a podcast with Paul the last time we had SPC-1 results and he really helped make sense of a lot of data - so watch out Paul, I'm coming for you when I'm back working next week. 

 

In the mean time, here are a few highlights and  I'll wrap it up with links where you can get more info.

 

  • SPC-1 IOPS = 258,078.23
  • SPC-1 price-performance = $.58/ SPC-1 IOPS
  • Average response time range for all ASUs:10 to 100% load =  0.33 to .86 milliseconds
  • Total ASU capacity = 1,145.324 GB
  • Protected Application Utilization = 70.46%
  • Total price = $148,737 (and I note that more than half of this is for the SSDs)

Look at the latency again - I'm sure there are a few jaws dropping with that.  The 7400 is a midrange general purpose array. What I mean by general purpose is that it supports HDDs, different tiers of storage and of course an SSD configuration - and these are very impressive latency numbers that more than demonstrate that the 3PAR architecture is not only for HDD but also ideal for flash. 

 

Get more details on the smoking hot HP 3PAR StoreServ 7400

The SPC-1 Executive Summary is available along with the Full Disclosure.  If you're not familiar with the SPC-1 benchmark, it might be useful for you to listen to the podcast I did with Paul Haverfield when we announced the results of the HP 3PAR StoreServ 10000 in October 2011.  There's also a blog post from Nate - who blogs at www.techopsguys.com that goes into a lot of detail so check out what he had to say. 

 

Lastly, I've started a series of ChalkTalks that focus on what I think of as the HP 3PAR "special sauce".  In part 1, I gave an overview of what makes the 3PAR architecture special and talked about the Mesh Active Cluster.  In part 2, I talked about the HP 3PAR ASIC and the controller node architecture so check them out. 

 

Now, back to focusing on my daughter's graduation!

Comments
Gchapman(anon) | ‎05-26-2013 03:27 PM

Do all customers get 45% off list price ? just just throwing that out there Calvin. For the record all the storage vendors give deep discounts on their kit for SPC tests, some  as high as 70% off list.  

diksha(anon) | ‎05-27-2013 03:42 PM

 

I can talk so much about storage now, I must say in which I have leant many things from your chalk talks.

nate | ‎05-29-2013 05:28 AM

Can I use this space to ask, beg HP to go buy some arrays from some of those other SSD startups that refuse to particpate in SPC-1 and put them to the test? The results would be very interesting to see.

 

NetApp did this to EMC several years ago, testing their CX3 I believe with snapshot support enabled and another test with it off. It was facisnating to see the EMC performance drop by something like 75% with that feature turned on. Especially coming from someone who has mainly used 3PAR the past many years, and there is no such performance drop (that I've ever noticed anyway) on that platform with snapshots.

 

 

| ‎05-29-2013 07:25 AM

@Nate - not sure what we have in our lab these days but I know a few years ago we had a competitor's array trying to reproduce their ESRP results which we could do but then noticed as we ran the test for a bit longer - the results dropped by 50% as their array fragmented. I'll leave it to readers to find the blog posts we did. 

 

That was interesting because the vendor ulimately claimed there must be something wrong with our array and they wanted us to send logs so they could look it over.  While we learned a lot (and I'm now always skeptical when ever I see any benchmark results from this vendor), not sure customers got as much out of it as we did.  There was a lot of he said/she said that I think made it difficult for customers to know what to think.

 

I've already sent your suggestion to folks on the team.

Josh S(anon) | ‎08-05-2013 12:25 AM

@gchapman - for what it's worth, I just bought two 7200's and got better than 45% off list on both. The spc pricing is definitely real world.

 

@CalvinZ - these numbers are very interesting... Emc recently claimed that their VNX 7500 has a maximum of about 170,000 IOPS. For a 7400 to smoke emc's top of the line VNX like that must be very satisfying!

Leave a Comment

We encourage you to share your comments on this post. Comments are moderated and will be reviewed
and posted as promptly as possible during regular business hours

To ensure your comment is published, be sure to follow the community guidelines.

Be sure to enter a unique name. You can't reuse a name that's already in use.
Be sure to enter a unique email address. You can't reuse an email address that's already in use.
Type the characters you see in the picture above.Type the words you hear.
Search
Showing results for 
Search instead for 
Do you mean 
About the Author
25+ years experience around HP Storage. The go-to guy for news and views on all things storage..


Follow Us
The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the Terms of Use and Rules of Participation