Should mainstream disk array vendors buy the flash startups, which are going to take their lunch money, now or later? Spend time with Kaminario, Nimble Storage, Pure Storage, SolidFire and Violin Memory, drink the Kool-Aid, hear them talk the talk and see them walk the walk. You'll think to yourself: "This is solid stuff, pure …
An interesting nascent market
Personally I think this market is interesting however is engineering for a Flash array really that difficult?
surely it shifts the bottlenecks away from spinning disk, so then you need to deploy your software, disk replacement and controller (a bag of processing, RAM, cache, I/O cards) more effectively so it can deal with the throughput. Stick in a few wear leveling algorithms and stuff around write caching and its just done?
The new companies that will succeed will be those that execute best on getting new customers and keeping them, making the biggest noise in the market and/or developing strong channels through which to sell it.
Flash may also turn out to be a disruptive technology, such as iSCSI, which has become prevalent, when once considered a game changer: by itself is now just one of the options for protocol choice when thinking about Ethernet-based storage, and a box-ticker for RFPs.
We shall see I guess...
Not sure about that, yet.
Hang on a bit, discs are not going away for a very long time.
Enterprise storage is in tiers, yes flash will replace tier 1, fast disc will move down to tier 2 and the cheap slow discs will replace the tapes in the lowest tier.
The big winners will be whoever can supply a fully integrated multi-tier solution, vendors with only flashy flash will only break in to that if they partner with existing storage vendors.
Takeovers would work with the established vendors folding the flash provider into their product but look out for partnerships as well as a possible future for these new companies.
Also the longevity of flash in an enterprise tier one situation where there are zillions of read/write operations a day may be questionable, the product is just not as mature and proven as spinning discs, granted the quality of flash is and will continue to improve so in time the longevity may be solved but it ain’t yet.
Not sure the big boys need to do that
I'm not so sure that the storage titans need to buy up the flash upstarts.
For example, Dell has Compellent, EMC has FAST, and IBM has SVC: all of these technologies do block-based tiering, and therefore integrate flash into storage arrays in affordable and effective fashion. NetApp is the odd man out here, with no such strategy, but that's intentional, I am told.
Granted, many of the new flash start ups are doing dedup to make the use of the flash effective, and none of the above vendors have dedup at fast enough rates for SAN. Still, I'm not sure that this is enough of a discriminator, to justify a buy, except perhaps to take the upstart off the market (which is illegal, but often done anyway).
Er, Dell and IBM "storage titans"? Hardly. EMC and NetApp are the big players, the HP's and IBM's of this world come next, then Dell just about warrant a mention and the rest are also rans.
why there is always someone compelled to make some comment about flash wear out in any article about enterprise flash arrays.
For large MLC arrays even if your wrote at full speed 24/7/365 for the life of the product you couldn't wear them out. I mean do you actually know anyone who writes the entire content of their storage system every couple of hours every single day? No you don't know anyone who does that, and if you did, well then they can buy the SLC version and write to it for decades.
I can't tell if the people who post "oooooo but it will wear out" in pretty much every flash related article are just ignorant or are shills for legacy disk array vendors, either way, it got old a long time ago.
Sometimes the need outweighs the costs.
If one of thier customers depends of high-performance, then these and even array bases solutions solve that problem.
If you look as a stock brokerage which may quantify its potential profits and losses as millions of dollars/pounds etc. per second, then the cost of these arrays pales in comparison.
At current hard drive pricing....
Flash is going to get a major leg-up. Enterprise or not.
I'm supposed to be trialling some SSD arrays shortly. it will be interesting to see how they stack up compared to the 90Tb spinning versions we've been using.