DSSD – Does This Give EMC the Tools to Lead in Flash?
In the technology world we see regular acquisitions happening. Just recently, a stealth startup named DSSD was brought into the fold by EMC which brought the eyes of many analysts and architects onto the company.
Without the financials being disclosed, it is difficult to tell just what value EMC has put on the line as they acquired DSSD. The press release states “The transaction is not expected to have a material impact to EMC GAAP or non-GAAP EPS for the full 2014 fiscal year.”. Of course, we have to understand that with a company the size of EMC with a market capitalization of 54.4 billion dollars, it is difficult to move the needle without a massive price tag.
What exactly is DSSD?
This is the question that will on the lips of every analyst who watches as the transition takes place. From what we know, DSSD is a privately held company with a technology referred to as “rack-scale flash storage architecture for I/O-intensive in-memory databases and Big Data workloads like SAP HANA and Hadoop.”
All buzzwords aside, this is a big gap for many storage vendors who aren’t necessarily targeting the large scale, big data implementations in an all-flash or flash-heavy deployment. Other vendors such as Pure Storage and Coho Data have distinct use cases that have been laid out based on customers they already have in production. The rapid scale out capability is a feature, but may not have their sights specifically set on “Big Data” targets.
Among the patents that DSSD owns, we can see that they have clearly put their engineers to work in creating some leading technology to target the shortcomings in massive scale out deployments on flash disk.
- Storage systems with incremental multi-dimentional RAID: http://www.patentgenius.com/patent/8341342.html
- Storage system with self describing data: http://www.patentgenius.com/patent/8370567.html
- Method and system for object-based transactions in a storage system: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=8601206.PN.&OS=PN/8601206&RS=PN/8601206
Welcome to The Family: Let me Show You the Coach House
Much like how EMC acquired XtremIO and retained them as a separately operating business unit, DSSD will be integrated to a degree but still operated as a standalone unit among the Emerging Technology Products Department.
There is a clear goal for EMC to take a role as a flash leader in the industry. Their sales show that with 17 petabytes, yes, petabytes in Q1 2014 alone. As an incumbent in enterprise storage, it is not surprising that they have the traction they have when they launch a new product. You may recall when McDonalds added pizza to their menu and instantly became the largest pizza chain in the world.
EMC, like rival IBM, is famous for purchasing companies with innovative patents and intellectual property, just to fold them under a business unit and let them languish before sometimes being launched into a product down the road. At the same time, there have been projects internally which did not bear fruit in the flash arena.
XtremIO stood the test of time and is being touted by its parent EMC as a strong part of the overall EMC flash portfolio. The question on everyone’s mind is whether DSSD will be granted the same fate.
Scale Beyond the Competition
As other flash storage vendors launch into the ecosystem, they have a variety of workloads that they will target, and specific reference architectures to attract customers. DSSD is no different, although they were scooped up before the full launch so we didn’t get to see the pitch in its full form.
Obvious targets for this would be large scale, high-volume databases such as Hadoop and SAP HANA. With the amount of analytics being run against retail, federal, and scientific research organizations, the ability to move those workloads into an efficient, scalable flash architecture is very attractive.
We will see how it all plays out over the coming months. The press release from EMC can be found here: http://www.emc.com/about/news/press/2014/20140505-06.htm and I think that we can look forward to some interesting work happening around this technology. The only question is how much, and how soon?