All-flash array startup Violin Memory has bought GridIron Systems, a startup supplying the TurboCharger SAN accelerating flash cache. Violin supplies shared 6000-series all-flash arrays to enterprises and the public sector as well as having single server PCIe-connected appliances and Fujitsu resells its array. It filed for an …
very surprised ...
I'm very surprised and see no synergy between Violin and Gridiron. Violin as a very good solution whereas Gridiron is juste another caching appliance, with the same revolutionnary approach as others ! I very much prefer Netapp one with their FlashAccel and the distributed way of caching, as closer as possible of the server!
The only interest i see is on a pure business Point of view. Gridiron could be dangerous for Violin so they could wand to stop them ! But what I beleive more, is that this is the first part of Violin of the plan ..... Next step will be that one of the Big Five will buy Violin.
And I think Netapp is the best for that: because Netapp are dead if they don't bring up soon a full flash, high performance, array. And it's too late to develop from scratch a new one. So they have to buy one ... And to be credible, they have to buy someone that is considered as a leader or a visionnary. Violin is. And Moreover, Netapp need his own caching solution, not the Fusion-IO one (or the FlashAccel one), because Fusion-IO will become very dangerous in the future, by entering in the Flash Array area. So Gridiron could be the Netapp way for building his own caching solution.
Lastly, Gridion and NEtapp already have strong relationship. Netapp E-series is Gridiron storage solution for their hybrid flash array ....
This is all why I beleive, and I hope, Netapp will (or has to !) buy Violin. If it's the plan, then the Violin / Gridiron wedding makes sens. It not, this Violin/Gridiron wedding is just a very sad and unanderstandable news for me !
I agree that this is an odd move for Violin since they don't need any caching due to being on the PCI bus when fully configured.
I don't follow why NetApp should want to buy the bundle though, they already cache to either NVRAM or SSD anyway and have a solution for moving closer to the CPU for reads. Their strategy also involves using SATA for the capacity behind the cache rather than faster disk so I can't see them planning this. I may be wrong though, these companies do some weird things!
Netapp needs something new, in the all Flash array area. Their FAS line of products are very good (IMHO) but for capacity needs. Nor performance needs. Their E-series line are good products too, with much better performance, but whithout any advanced feature. And they still can't compete with Full Flash actors.
It's why they have to push out a new solution, either custom made or thanks to shopping cart !
This is what is behind the confidential "Mars Project". And the timeline is probably arround Mars 2013 ... So this is why I think that Netapp might be interested in buying Violin.
Moreover, everybody know that Netapp like "All in one " Box, with SAN (FC, ISCSI, ...) and NAS (CIFS, NFS) features. And in november 2012, Netapp secretly bought CacheIQ company , which similar feature as Gridiron, but in the NAS area...
So if Netapp Buy Violin, they will have all the necessary technologies to build a very interesting full flash solution: Violin for the Array, Gridiron and CacheIQ for the cache (so they could stop using Fusion-IO technology), and their own technologies for everything else (management, dedup/compression, ...).
Of course, this is all pure speculation from me. But I think It makes sense, in a technical and business point of view.
still very surprised !
On january 16, Violin wrote that "caching" is a compromise, totally useless in their point of view: http://www.violin-memory.com/blog/the-cost-of-compromise-4-caching/
And on january 18, Violin CEO say they buy a caching solution .....
So this sound still, and moreover, a very surprising news ... To strange for me