The rumors have been confirmed—EMC (NYSE:EMC) is indeed acquiring XtremIO. Like Pure Storage, XtremIO is crafting a next-generation storage array, purpose built for 100% solid-state flash. Gartner recently forecast that this all-flash array market is going to reach $4B by 2015. It’s not at all surprising that EMC wants a piece of the action. Our expectation is that the XtremIO technologies will eventually get folded into EMC’s Project Thunder (announced earlier this year for delivery in 2013), and/or serve as the basis for an entirely new all solid-state array.

At the same time, EMC has already shipped more data center flash within their disk-centric arrays than any other vendor. The fact that they are acquiring a competitive architecture for their future all-flash technologies is indicative of something interesting going on. We can draw two conclusions from EMC taking this action that will have a broad impact on the storage market:

  1. Storage architectures designed for mechanical disk are not a fit for all-flash. We’ve long believed that taking proper advantage of flash requires a comprehensive redesign of the array hardware and software. (For more on this, see our Top Ten Reasons Why Flash is Different.) For years, EMC has been selling flash cache and tiers as performance accelerators for their CLARiiON/VNX and Symmetrix/VMAX products. With both XtremIO and Project Thunder, EMC has endorsed the view that all-flash storage demands an architectural rethink.
  2. Deduplication is key for all-flash.  XtremIO was pursuing a solution like Pure Storage’s that uses dedupe to both lower cost and increase performance.  Because flash writes are relatively expensive and random I/O cheap, it does not make sense to write the same data over and over again to flash the way it is done on performance disk. With Pure Storage’s launch in 2011, the industry got the first proof that deduplication (and compression) can be accomplished inline with 100Ks of IOPS and sub-millisecond latency. It seems EMC agrees, as buying XtremIO is a bet on dedupe algorithms crafted specifically for flash. (For a more detailed analysis of impact on the broader storage market and our proposed emerging segmentation, please see this post.)
We have remarked before that disk is the new tape and flash is the new disk in that from the perspective of a CPU doing random I/O, disk today appears slower than tape did 20 years ago. It is interesting, then, to consider last decade’s secondary storage media transition from tape (the slow media) to disk (the new faster media). In the first phase of the market, various attempts at HSM tried to mix tape and disk in the same online file system, but complexity and hugely variable latency caused most of these data stores to eventually go all disk.  In the backup world, VTL enabled the adding a performance-enhancing “disk cache” in front of large tape silos, but again the all-disk approach prevailed with DataDomain’s launch of a solution that was purpose-built for disk and that relied upon deduplication (and compression) to bring the cost of disk inline with the cost of tape. These are precisely the trends EMC just reinforced for flash storage, and they run counter to their focus on selling hybrid arrays.
We believe history will repeat itself in the all-flash market. If you want to accelerate the transition from a slow, cheap storage media (tape, disk) to a faster, more expensive one (disk, flash), find a way to take cost out of the equation. Pure’s secret sauce is that we are delivering a product that is 100% flash, but doing at $5-10/GB usable, a price at or below what most customers are paying for performance disk. The good news for us for now is that there are still substantial hurdles in front of those that would strive to replicate our unique recipe.


What does this mean for Pure Storage?  Pure Storage customers and partners will wake up tomorrow and realize not much has changed. Of course, we’re gratified that the all-flash design we have been shipping to customers for more than two years has now been validated by the storage industry leader. And no doubt our already high sense of urgency is going to be just a bit higher going forward. EMC is already one of our primary competitors. (As the XtremIO product seems to have only recently begun Alpha trials, we have not yet had the opportunity to joust). We welcome that competition.  Competition drives innovation and maximizes customer value. So long as we can convince potential customers to give the FlashArray a try, we are confident we can continue our winning streak.

I love this Steve Jobs quote from Steve Lohr at the NYTimes: “In a conversation years ago, Jobs said he was disturbed when he heard young entrepreneurs in Silicon Valley use the term ‘exit strategy’ — a quick, lucrative sale of a start-up. It was a small ambition, Mr. Jobs said, instead of trying to build companies that last for decades, if not a century or more.”

We practice that mindset at Pure Storage: we remain convinced that the best path forward for all of our constituencies—customers, partners, employees, and investors—is for Pure to continue our exponential growth as an independent company.  After all, the market is now coming to the playing field we staked out back in 2009. Moreover, there’s something subtle that happens to an entrepreneur when you start to focus on selling your business: you take your eye off of Job’s ultimate prize—changing the world by building a quality product and business that stand the test of time. Today’s performance storage customer isn’t getting the value they deserve for their money, and as in independent with great partners, we’re in a unique position to lead the industry transition from mechanical disk to solid-state flash.

It is shaping up to be a wild decade in storage, but the biggest winners will be the end users who get to unshackle their applications and administrators from mechanical disk, and save money while doing so!