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

With the All-Flash HP 3PAR, no reason to go to Xtremes

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

With all the anticipation (and delays) around EMC’s announcement of XtremIO, their all flash-array based on the $430M acquisition of XtremIO last May, I wanted to recap what is being said and look at our HP 3PAR StoreServ 7450.  But I do want to start by giving you an overview of the 7450, so you know where I'm coming from.

 

A Quick Look at the All-Flash HP 3PAR StoreServ 7450

Prior to announcing the all flash HP 3PAR StoreServ 7450, Priyadarshi (Pd is the product manager) did several blog posts talking about the technical requirements for an all-flash array.  and an HP Tech Talk where he did a bit of a deep dive into the all-flash 7450 with video blogger Andy McCaskey and storage expert/blogger Nigel Poulton asking him questions.

 

 

 

 

One more blog I recommend you check out is Revisiting the HP 3PAR StoreServ architecture advantages for flash.  One of the key advantages we have with HP 3PAR StoreServ is that it isn't a single silo array architecture - this blog hones in on that. HP 3PAR is a single architecture that scales from the low-end of midrange arrays (starting at $25K) through enterprise Tier-1 (up to 2.2PB) and in the middle is the all-flash 7450 (when performance matters). 

 

The Basics of XtremIO

I’ll start with a basic look at XtremIO from The Reg’s Chris Mellor.

 

The basic building block is called an X-Brick.  It’s 6U and EMC is stating 250,000 random 4K read IOPS and sub-millisecond response time.  An X-Brick has 25 X 400GB eMLC SSDs, for 10TB raw and 7TB usable.

They can “scale” to 4 X-Bricks – more on that in the “My Take” section.  Two bricks take 12U of rackspace and four bricks take 22U – I think the difference is related to either battery backup or Infiniband switches for communication between bricks.
 
A brief look at software reveals integrated workload and data balancing across SSDs and controllers, thin provisioning, inline cross-cluster deduplication, and VAAI integration.

My Take

Despite a few people out there saying that this is a second generation product, it isn’t.  EMC has just announced the GA (generally availability) of XtremIO. Having it in a few customers hands doesn’t make for a battle tested product.  And from the things it is missing, it clearly is a first generation product.  Here’s what I see:

  • Software missing from the offering is any kind of Quality of Service, encryption, snapshots or clones.  There was no mention of federation.  XtremIO won’t natively talk to a VNX or VMAX so it’s the 5th silo in EMC’s multi-silo array portfolio.
  • EMC also said during the announcement that you cannot scale an existing XtremIO array – so if you start with 1 X-Brick, you can’t upgrade it to 2 or 4 X-Bricks later.  I’m not sure how it can be scale-out if you have to buy into the largest configuration you think you’ll need at the start. They have a small, medium and large configuration without an upgrade path. 
  • A 4-Brick XtremIO would require 8 dual socket CPUs (2 per  X-Brick) that is a lot of CPU power for 1M IOPS.  From a rack density perspective, they are about 1.8TB per U of rackspace (50TB / 22U).  The HP 3PAR 7450 has 11TB per U of rackspace.
  • EMC claimed they are the only all-flash array with full VAAI integration.  I guess they forgot about the 7450.  They also didn’t mention VMware MetroCluster Storage support which is supported by 3PAR and not XtremIO. 


Robin Harris at StorageMojo

A post worth reading is “XtremLY late XtremIO launch next week” from Robin Harris that he obviously wrote prior to the announcement.  Here are a few things Robin said in his post:

  • EMC’s XtremIO product is launching next week on November 14. This is the beginning of the end for EMC’s Symmetrix line and the toughest product transition they’ve ever attempted.
  • Veterans know that a long beta is a sure sign of a troubled program. But let’s put all that behind us. Where are we 11 months later?
  • Despite a year’s worth of effort the XtremIO product is not as mature as ones shipping for a couple of years. EMC is caught in a bind of its own making – which doesn’t happen very often

Bert Latamore is less than overwhelmed

Ten minutes after posting this, I saw another blog pop up titled EMC's long-awaited XtremeIO flash array: less than overwhleming.  Bert wrote for SiliconAngle.com a few interesting things:

 

  • Overall the initial announcement is less than overwhelming, considering it comes from the dominant storage provider, a company focused solely on storage. Both IBM and Hewlett-Packard, which have had all-flash arrays with greater capacities and competitive performance on the market for several months...
  • The XtremIO system is definitely a good solution for some applications, as demonstrated by the eight-months of production trials by EMC customers. At the same time it is not an industry leader, and this is not the announcement that the industry looked for from the market leader. Is it enough to protect EMC’s dominance in the storage market? Maybe, but EMC has definitely left the door open here, and both IBM and HP are committed to increasing their presence in the storage market, largely at the expense of EMC.
  • EMC has to stop worrying about protecting its existing empire of disk storage and realize that it is in real danger of becoming a legacy vendor in the new reality of flash. As the price of flash drops and its advantages become compelling for increasing numbers of business use cases, it needs to be fully competitive with IBM, HP, and the all-flash array vendors to hold its position.

 

Final Thoughts

There's no doubt that flash technology is a game changer.  Our technology can tend to overhype new technology - flash isn't being overhyped. And it's great to have the storage market share leader EMC finally with an all-flash array.  I do find it interesting that EMC started hyping SSD's back in 2008 and said that SSD would be at price parity with Fibre Channel (spinning) drives by the end of 2010; they obviously were a bit off with that prediction.  Given the early hype of SSD, I find it ironic that they're almost last to market with an all-flash array. 

 

I think the portfolio we have with HP 3PAR is as strong as what anyone has in the industry.  Any customer looking at midrange, tier-1, or all-flash storage should be taking a hard look at HP 3PAR.

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
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