With $11bn at stake, you'd think that someone would have solved the virtualized storage problem by now. The problem, however, is that there's so much more money in the existing storage market. It's an "innovator's dilemma," Clayton Christensen style. It's therefore not surprising that it took a newbie Tintri to try and up-end …
Storage isn't a barrier to virtualisation and hasn't been for some years. Where did you get that notion? Perhaps from some PR officer.
No, actually, for most enterprises it's application and OS vendors who are the problem, usually through bogus/archaic licensing or support certfication constraints. Not technology problems at all.
Sir (Matt Asay),
the term "virtualized storage" actually means something else entirely than the used meaning of "storage for vitualized systems".
So again, I have to say that the IT Field of Storage is something not just about anybody can fiddle with. Taking your bootnote into account ("Matt Asay is senior vice president of business development at Strobe, a startup that offers an open source framework for building mobile apps") I would have to say that you either have no technical knowledge of storage or that your technical knowledge of storage applies to either the opensource software field or the mobile app field.
Having had a brief look on tintri's website, the problem with tintri's offering is that it is highly dedicated and nowhere near a brandnew idea. EMC's VMAX with Enginuity 58.75 (for example) can do exactly the same thing for many more application behaviours than tintri. tintri is thusly a niche player targetting the lower end of the storage market. Not really a market up-ender in the semantics that you proclaim.
Sir (Joshua Goodall),
I fully agree with your sentiment, and we both know that storage vendors actually try very hard to make storage virtualization a real reality (IBM's SVC, EMC's VPlex, HP's (rebranded) MPS200, Brocade's Application Module for which they sell a Data Mobility Blade to name just a few).
Having been around the block for nearly 12 years now, I must say that storage still is a huge barrier to virtualisation. Countless SMB's either use internal storage or a smallish mostly NAS-based diskcontainers as storage. True, most of those SMB's don't really have virtualization requirements, which is reflected in the 70% of enterprises that do not or did not yet virtualize their services.
There is also no offering available currently, that allows an SMB to outpace server-slice growth with storage growth: Hosting providers offer a direct link between amount of GB disk and amount of CPU&RAM.
Configuration of Storage?
Not sure if what you mean was more the configuration of storage, or creation of dynamic storage units? Complete storage abstraction?
A perfect example of the supported / not supported vendor situation in the past was the Exchange fiasco with VMware and Microsoft. This type of problem occurs when vendors do not have unified direction for QA/Support/Engineering, or perhaps nepotistic interests. VMware owned by EMC is a decent example, but I think that the entire ecosystem is a bit more complicated. This is exacerbated when vendors build other vendors' software into their own. I personally watched a virtualized product on super fast hardware and storage have a slight latency in the indexing component that cascaded into a real mess over time. If you don't verify the write, or cannot verify the write, then you open the door wide for corruption...or the application thinks that it failed. How do you abstract that? Worse, perhaps it did fail, and the application thinks it was successful. Database <-> application mismatch, indexing corruption, misplaced entries, metadata problems, on and on. Then QA/Support/Engineering says, "It was not tested we are not going to support it."
If these guys have solved the problem, then I would say that they are in the Buzz phase at the best. I personally would not be implementing or recommending implementation.
Was there a point here? This reads like someone trying to use the internet to get their buddy's company sold. How about some actual tech related information. This article was just crap vomited out.
- Pic Mars rover 2020: Oxygen generation and 6 more amazing experiments
- Microsoft's Euro cloud darkens: US FEDS can dig into foreign servers
- Plug and PREY: Hackers reprogram USB drives to silently infect PCs
- Review Fiat Panda Cross: 'Interesting-looking' Multipla spawn hits UK
- Analysis PEAK LANDFILL: Why tablet gloom is good news for Windows users