Tegile’s claims are so impressive that they scream out that something must happen to the company and to those like it. Tegile reckons that it can provide, for example, 105 TB of storage at 75,000 IOPS in 2U of space for $75,000 against equivalent I/O and capacity from NetApp that would take 115U and cost $475,000.

Tegile gets the impressive results it claims by a combination of DRAM cache, MLC and SLC storage tiers, and SAS drives. This is all cooked together with a ZFS-based OS tweaked by Tegile to provide data deduplication; compression; RAID enhancements; and a performance-boosting feature called MASS, or Metadata Accelerated Storage System. MASS effectively means that data, once ingested into the Tegile system, is dealt with pretty much via just the metadata headers rather than the full copy, and these are kept in cache or SSD tiers. See this article by Dave Raffo for a fuller rundown of the spec.

And in another example given by Tegile vice president for marketing Rob Commins, the company provided customer Virgin America with the kit it needed for $80,000 and saved it from spending $870,000 with EMC.

Such cost, performance and space advantages are rather disruptive to say the least, and it will be interesting to watch the reaction of the big vendors.

Were the world of storage marketing and sales a level playing field, the likes of EMC and NetApp would be in big trouble. They wouldn’t be able to stand a competitor offering equivalent performance at 10% to 20% of the cost of their products. But it isn’t. The power of the incumbent is huge in terms of marketing and sales, as is fear of the unknown among customers.

Of course, the likely outcome is that these innovative startups will be swallowed up by the big vendors at some point, as noted in my ‘Storage and capitalism’ blog post. Not because they pose a great threat – they’re the bite of a flea in terms of revenues diverted from the big boys – but because EMC or NetApp want their technology (or don’t want a competitor to get it).

We had an example of that recently, with EMC’s acquisition of XtremIO, another hybrid flash/disk/dedupe startup, for $430 million. EMC has incorporated it into its Project X, which aims at providing high-IOPS storage for virtual environments.

EMC is the storage market’s giant, and it surely has made the first move in a likely round of consolidation. It’ll be interesting to see who gets bought by who, how startup technologies are incorporated into big vendor roadmaps and who gets left unacquired or bought and left to wither on the vine.

Follow me on Twitter: AntonyAdshead

Perusing this week’s selection of content on SearchStorage.co.UK I am spurred to remark how the business of storage reflects the economy/society we live in. A few key facts of capitalist life pop out from the stories we feature.

Fact No. 1: Economic power consolidates and big companies tend to lord it over small ones and eventually absorb them or wipe them out. See Chris Mellor’s argument for why the big storage vendors will sweep the board in flash cache. Also see Symantec’s attempt to chisel some cash from (and fatally wound?) Veeam and Acronis in a lawsuit filed last week.

Fact No. 2: The nature of private enterprise builds in duplication of technologies while also ensuring similar products are incompatible. This occurred to me when writing the Greenbytes blog post and the fact that their product won’t work with any other array, or at least not without storage virtualisation tools. I’m not singling out Greenbytes here – they’re no different to any other vendor; they all make sure you can’t use their arrays with other vendors arrays. Lock-in is the name of the IT vendor game in capitalism.

Fact No. 3: Periodic technological advances threaten entire existing paradigms. Here I have in mind the story showing how cloud file storage services are undermining VPN use and even in-house file servers. The cloud has the potential – once security and latency concerns are dealt with – to severely dent the existing model of in-house storage arrays. Economics of scale (see Fact No. 1) will likely see cloud providers being able to provide far more storage far more efficiently to many more people than are catered for by existing modes of storage.

Follow me on Twitter: AntonyAdshead

At the morning keynote Darren Thomas, vice president and general manager of Dell’s storage division, sought to underline that confidence with a narrative of Dell’s transition from EMC reseller to storage vendor with its own technology.

Thomas expounded a vision of the future of “fluid data” and a seamlessly integrated product portfolio. At one point he told the audience of a Texan aphorism — “big hat, no cattle” – when making reference to competitors.

But what exactly does Dell have in the storage stockyard right now?

These are the key technological components of the Dell storage roadmap, based on the concept of “fluid data” as outlined by Thomas:

· “Dynamic tiering”

· “Better together”: a reference to the goal of developing an integrated product portfolio, with interoperability between and using Dell’s key acquisitions (EqualLogic, Compellent, Exanet and Ocarina)

· “Self-protection”: data protection “at the push of a button” to supercede traditional backup

· A “consistent ownership experience”: see “better together”

· “Native cloud”: a recognition of the cloud as a key storage tier

· “Scale-out designs”

Looking at these in detail, what emerges is a roadmap rather than a current reality.

Tiered storage, for example, exists in Dell’s Compellent products. In fact, Compellent was a pioneer of strorage tiering. But, that said, there’s no way of tiering across products, say between EqualLogic arrays and Compellent. There’s also the issue of the flash tier, which we’ll return to below.

Dell fares better on its “better together” goal, although “nearly together” would be more accurate. Its PowerVault and EqualLogic product lines both got the Fluid File System (which resulted from Dell’s acquisition of Exanet) last year, which allows file-based access to these subsystems. Compellent will get it in 2012 sometime.

Dedupe/compression (from the Ocarina acquisition) is already in the DX object storage platform and the new DR4000 disk-to-disk backup product and will be incorporated into the Fluid File System, so will get into Equallogic and Compellent arrays that way.

Meanwhile, Compellent products are set to get a 64-bit makeover on their controller OS. That’ll mean massively increased addressable block numbers for one thing, but EqualLogic won’t go 64-bit, and those two products will retain separate OSes. Thomas expressed a general aim to get Compellent and EqualLogic arrays to work together more, such as allowing replication between them, but that’s also a work in progress. So much for “better together” and “consistent ownership experience” there.

“Native cloud” is also at the aspiration stage right now. Dell’s storage vision is based on the idea of tiers, and the cloud is one of those recognized tiers. It’s just that right now no products allow tiering to the cloud, but they’re “working on it,” said Brett Roscoe, general manager and executive director for Dell’s PowerVault and data management solutions.

When it comes to the idea of tiers, right now we need to talk about server-side flash, the arena where the likes of Fusion-io have made waves, driven by the huge random I/O needs of server and desktop virtualisation.

Dell was coy about it today. Thomas talked about the potential pitfalls of large chunks of data being in server-side flash and a failure happening and the need to link many such lumps of memory together. “Flash without resilience is no good,” he said, adding that “there’ll be an announcement … sometime.” This is clearly a reference to the technology acquired from RNA in June last year, which includes server memory acceleration.

So, summing up, I wouldn’t say Dell is a “big hat, no cattle” vendor. It certainly has a big hat, and it has cattle. It’s just they’re a bit of a way from all being the same breed right now.

Two recent emerging vendors/products have cited the demands of applications, virtual servers and desktops as drivers for the location of storage right next to the hypervisor. The first is Nutanix, which has come up with a sort of clustered DAS.

The second is Fusion IO, which I spoke to this week. Fusion IO sells NAND flash that goes into the hypervisor server, acts as local storage and takes the load off arrays that struggle with the huge, random I/O demands of multiple virtual machines.

Fusion IO has shipped 20 PB of its stuff and, among other things, claims to totally remove the dreaded “boot storm” of VDI environments, in which multiple users start their desktops at the beginning of a work shift.

So, it seems the needs of virtualisation are driving a move to locate storage near to the CPU so as to reduce latency. Whether it all needs to be there is the question; perhaps it’s just a super cache Fusion IO-style that needs to be so close to the action?

PS: There seems to be an unspoken axiom in storage along the lines of “hardware is fundamental, but software is where you make money/differentiate your product.”

As if to illustrate this, Fusion IO, a relative newcomer to the market, last year launched the ioMemory Virtual Storage Layer (VSL). In doing so, Fusion IO moved from simply being a company (and not the only one) that sells server-based NAND flash to one whose software could now help that flash operate way closer to the server’s memory than a simple bolt-on flash card, further reducing latency in the process.