Screenshot 2014-12-05 15.56.38I’m still in the U.S. for the last day of the IT Press Tour, and the prize goes to Primary data for the best meet up of the week, by far (no matter what I see next)!
I spent only an hour with them and I’m now very thrilled and puzzled at the same time. Best vision and ideas I’ve seen in years but, at the same time, a vision that is incredibly hard to achieve. Everything is so immature that only faith, boatloads of money and the perfect strategy (all together) will lead this company to success.

Thinking about Primary data

Already $60M in funding, an A+ team of storage personalities, a lot of marketing (with all due respect, I can’t think any different when guys like Steve Wozniak are involved), three R&D sites, out of stealth mode only after one year from its inception, with a bunch of PoCs and a (probably totally immature) product out very soon. There are only two possibilities here: either these guys know exactly what they’re doing or they are totally out of their minds. (I’m for the first one… but, you know, VC funding could easily help to shift towards the second one).

Thinking about Nicira

Screenshot 2014-12-05 15.56.07First things first, just speaking about marketing, Primary data is very adamant about this: “this is not a software-defined product”. From my point of view it is, but it is so advanced (and better than anything we are used to) that talking about SDS will only diminish the value of the proposition.

The best way I know to describe Primary data is to compare it to Nicira. Nicira was acquired from VMware for $1.2B a while ago, is a pioneer of SDN and its technology is now at the base of VMware NSX. I’m going to borrow some of the basic concepts behind NSX to explain what Primary data does even if in most of the cases it is an over-simplification.

As for Nicira/NSX we have three major components: the protocol, the virtual switch and the controller. Translating this in Primary data words we have: pNFS (comparable to Openflow), the primary data hypervisor (the Open vSwitch), NSX controller (the Data director).
Also the functionalities are very similar:

Screenshot 2014-12-05 15.56.27– all the storage virtualization happens in the data hypervisor, which is installed in each single server. This is a very complex piece of software indeed, it is a kernel driver and also does al the protocol conversion (file/file, file/block, file/object)
– pNFS “encapsulates” the traffic and it is used for the transportation between data hypervisors. (there is much more complexity here but you should first understand how pNFS works to go deeper, and this is not the aim of this post)
– the Data director is a policy-based metadata controller which is in charge of all data positioning, moving, etc. (as for Nicira, all the magic, the potential and the money is here!!!)

Like for Nicira and Openflow, if the storage vendor supports pNFS natively (and seriously), I’m sure there could be some magic here… but at the moment it’s only speculation.

There are other important pieces in Primary data architecture but, at the moment, I think this is enough to give the basic idea.

Thinking about reality

The data hypervisor is a kernel module (a device drivers) and, as such, is something you can implement only if you have complete support from whoever controls the kernel.

It’s not a major problem for Linux, if I understood well, Primary data will be releasing the data hypervisor to the open source community (like Nicira did), this would contribute to its adoption and, in the medium/long term it could become part of standard linux distributions.

Things get much harder when it comes to other operating systems or hypervisors like, for example, in the case of VMware. In fact, I can only imagine the reaction of EMC to a similar product… Do you remember? Also Nicira product was much more advanced on Linux than VMware before acquisition.

At the same time the product is still in beta, very immature and probably limited. Even if the company is already out of stealth mode it doesn’t have an “off-the-shelf” GA product that you can buy and install tomorrow… this will take some time.

Closing the circle

Primary data has a huge potential and a totally disruptive technology in its hands. But it is too early to say if, or when, they will be able to exploit all of this.

It’s quite clear that they will be getting a lot of attention in the next future and anyone who makes or deals with storage will definitely be keeping theirs eyes on them.

I can’t really see Primary data becoming an important storage company but, if I can dare to make a prediction, they will probably be bought for an amazing amount of money from a big vendor sooner than we can expect…or they’ll die out because no one in the storage industry will support them (they are deadly disruptive for traditional vendors!). The first prediction is much more likely of course, and I would bet my money on VMware or, why not, companies like CISCO.

Disclaimer: I was invited to this meeting by Condor Consulting Group and they paid for travel and accommodation, I have not been compensated for my time and am not obliged to blog. Furthermore, the content is not reviewed, approved or published by any other person than the Juku team.