Doing It Again: EMC XtremIO or Pure Storage?

A fellow technologist asked a very fair and controversial question in a comment to IOPS Matter: VMware Native Multipathing Rule Attribute Affects Storage Failover, which pertains to my VMware-XtremIO environment. Since my response was running quite long, I thought it better to re-post the question here, followed by the answer.

“We are looking at purchasing a new all-flash SAN for our SQL environment running on VMware 5.5 — in your experience between Pure and EMC XIO, if you had it to do over, which would you buy? We are looking at the X-Brick 10TB against the Pure FA-405 6TB models. SQL compression is about 1.7:1 and dedup is almost nothing until we talk about storing multiple copies of our 300GB database for dev, test, staging, etc. Other than consistent finger-pointing from vendor to vendor, I’m not seeing much difference that would concern me in either direction other than price and that Pure’s 6TB might not exactly match the 8-9TB available in the XIO. Feedback?”

That’s quite the question, the answer to which would become headliner marketing material for whichever product was endorsed. Thankfully for me, the politically “safe” response of “it depends” is actually true. Factors like price, observable data reduction, and I/O patterns all sway the arrow.

Personally, I dislike the price factor because it biases the more objective technical virtues of products. In dollars, though, my experience has shown that EMC is willing to go the distance to make sure they aren’t outbid by Pure. That probably varies by VAR and their relationship with EMC, but EMC’s vast empire and capital make them hard to beat in this realm.

It sounds like your environment is similar to ours in data type (we’re heavy MSSQL, too). Thus, I’d expect to see comparable reduction rates for you as for us; it could vary depending on what is stored in your SQL databases, but let’s assume they are close. On Pure, we saw around 4:1 data reduction, due to commonalities in databases and dev/test/stage environments. That’s *reduction*, not just deduplication. We don’t yet know what that ratio will be on XtremIO, because we’re still on 2.4 (3.0 just GA’d and isn’t seamless). Thus, if you have to pull the purchase trigger before we or another XtremIO customer can post real-world compression results, you’ll be comparing “what is” to “what is claimed to be”. I don’t have a good history with that comparison, so if you can wait and see, I would.

The last part, I/O patterns, is related to data reduction because you have to favor one. It’s physics. Pure Storage uses variable-length block deduplication, which takes more time to do, but which also achieves better deduplication ratios (see the strength of EMC Avamar for proof of that). The downsides are post-processing and garbage collection, which EMC Marketing is proud to highlight. XtremIO, on the other hand, uses fixed-length block dedupe (4K in pre-3.0; 8K in 3.0+), which makes it fast to process on the fly (read: no post processing or GC), but which doesn’t achieve the same degree of dedupe. Both of those are physics choices, architecture choices. Personally, I don’t think it’s a good vs. bad comparison; it’s more apples and oranges. Both are fruit but they peel, slice, and scale differently.

Back to the practical in I/O, our patterns required some tweaking of Pure’s deduplication processing intensity in order to maintain low latency. The processing weakness of variable-length dedupe is large block sizes. Backups use large block sizes, and for better or worse, our developers choose to run full SQL backups all at the same time each day, which gave that dedupe processing a run for its money. With help from Pure support, that behavior was mostly mitigated. We didn’t see any other challenges with the handling of our I/O patterns. XtremIO doesn’t flinch with that same backup flood, but it also doesn’t touch even half of the data reduction rate (pre-3.0). Thus, they are *very* different beasts pre-XIOS 3.0. I hope to find out in the next month how they measure up in the 3.0 world. Granted, we don’t have a Pure FA-400 on hand to directly compare with, but our environment hasn’t significantly changed since we gathered our data in late 2013.

I’m including a link below to EMCer Chad Sakac’s blog entry on XIOS 3.0’s disruptive nature. I’m not a big fan of parts of the justification in the post itself, but I found that the dialog in the comments was useful information. You’ll see a good mix of support and push back in the replies. Take a moment to read them and sample some of the emotions swirling around the All-Flash Array market. The replies from current EMC and XtremIO customers are particularly applicable.

http://virtualgeek.typepad.com/virtual_geek/2014/09/on-disruptive-upgrades.html

Wrapping up, we both know that I could be more definitive in answering the question, but I haven’t. Part of that comes from the fact that I’m biased by some experiences we’ve had (with both products, actually) that are now in each product’s history. It wouldn’t profit you for me to warn you about situations that were addressed in late 2013 or early 2014 software releases, because you won’t see them (unless you figure out time travel and return to repeat them). The other part comes from the current absence of data parity–the “what is” versus “what is claimed to be” comparison.

And that’s the twist. The all-flash storage and data reduction spheres are all about the future, so you would be remiss if you didn’t pay attention to what vendors claim and project for their product road maps. The key is finding harmony between “what will be” and “what is” today. If you do decide to go with a product based on its claims, make sure you get it in writing that the product will deliver on that claim. Worst case, you return it and go with the alternative. Best case, the vendor delivers what it takes to meet the claim, which means bonus hardware. Either way, you gain experience and hopefully only lose time.

Next up: Doing It Again: How Would I POC XtremIO and Pure?

10 Comments

  1. Michael Chouinard said:

    Have you upgrade to 3.0, yet? I’m very interested to hear your impressions.

    November 20, 2014
    Reply
    • Chris said:

      Hey Michael. We currently have a test/temporary X-Brick that we’ve been using with our NDU case to repro the failure there (if you have QLogic CNAs, make sure RR IOPS=1), and that brick has been upgraded to 3.0.

      The upgrade process took a little over an hour with a remote engineer driving. Since it wipes out everything, we’d already deleted the volumes and the resulting upgrade basically presents a clean box, as if you just received it. We didn’t have any initiators created at the time, so I mapped them after the upgrade and created volumes.

      We’re still seeing how it handles the data–and a full post will likely follow about that–but thus far it is a bit meager. We have a couple sets of backup data on disk that we know don’t dedupe well and having them on 3.0 garnered a whopping 1.0:1 ratio. We had to roll that over to spinning storage after we realized that, because the rest of our stuff won’t fit.

      I’m working on an EMC Avamar case that has my migration temporarily paused, but I hope to get back to it either tomorrow or Monday. Then we’ll see how it handles SQL data. I’ll post some stats after that. Being that we don’t use anything beyond block storage, 3.0 is the same as 2.4 to us, plus compression. More to come!

      November 20, 2014
      Reply
    • Chris said:

      Michael, this reply may be irrelevantly late in coming, but I realized that I hadn’t provided any “after-3.0” results. Unfortunately, XtremIO didn’t deliver on data reduction hopes with 3.0. Before 3.0, we saw 1.5:1 deduplication. On 3.0, deduplication actually dropped to 1.3:1, but compression added 1.3:1, all for a net total of 1.7:1. Thus, we only saw a negligible gain from the upgrade.

      Apart from that, 3.0 is a little fatter on the XMS (XtremIO Mgmt Server) appliance and now uses a nearly-900GB thin provisioned paravirtual SCSI disk. Apparently it will use that for some future feature. The rest of the features weren’t a value-add for us.

      March 19, 2015
      Reply
      • MIchael Chouinard said:

        Chris,

        It is true that the reply was a bit late in my decision process, but thank you nonetheless. We ended up purchasing XtremIO for our purposes, which was to combine our entire environment on to one array.

        We looked at XtremIO, Pure and Tegile. We were not in a position to do PoC, so we relied heavily on posts like yours, references and trusted partners. We quickly came to realize that we were going to get the performance that we needed from any of the solutions, as my environment will never demand anywhere near 100K IOPS. In the end, we went with the solution presented to us by our most commonly used partner… we were also swayed away from one vendor by the nature of their sales team.

        Having not done a PoC, EMC ran a scan to determine what kind of data reduction we could expect to see. They stated that if we did not get the promised results, they would “fix” it. Our VDI environment was a bit different that they had seen before and our real world reduction was quite a bit different than the scan. I can say that EMC actually stood by their promise and is working on a resolution for us.

        Michael

        April 3, 2015
        Reply
        • Chris said:

          Hey Michael,

          Thanks for the update. With the quality of the leading enterprise products out there, it’s more of a “better and best” decision than a “right and wrong” one. EMC is good about making sure the solution delivers, even if that means the solution grows by several rack units :). It went the same with us.

          That’s unfortunate about the one product’s offensive sales team. I’m always disappointed when a promising product gets impeded by folks who aren’t seeking the customer’s best or are otherwise an obstacle to the solution.

          Hope your flash future continues to go well! Shout out any time.
          Chris

          April 3, 2015
          Reply
  2. David Czech said:

    Chris,

    Thank you for your informative posts.

    I would be very interested in hearing any additional details and updates you would would be willing to share about your comparisons between Pure and ExtremIO, especially as you have had the ExtremIO in your environment for a while now.

    If you could, please also give some insight on how you chose those two as the best competitors for your environment.

    April 2, 2015
    Reply
    • Chris said:

      David,

      I’ll be posting an update soon, as we just went through a fresh RFP process with EMC XtremIO, Pure Storage, and HP 3PAR and selected Pure as the successor to our outgoing 3PAR V400. This means we’ll have XtremIO and Pure at sister datacenters. It should be an interesting elaboration when I’m able to flesh it out. I’ll add a link here so you get pinged when it publishes.

      Thanks!
      Chris

      April 2, 2015
      Reply
  3. Guy said:

    Also trying to decide between Pure, Tegile, EMC and Violin… All would be enoigh for our needs, costs are one, trust is another… Pure seems to have best team spirit, but not an easy decision to make… Will be watching your upcoming post(s) :)

    April 11, 2015
    Reply
    • Guy said:

      Thanks!
      Maybe we will get a Pure and an XtremIO Brick, EMC is currently bundling Recoverpoint for free which is a nice thing :)

      April 22, 2015
      Reply

Leave a Reply