The device – called the Solidarity — offers a dual-controller unit with inline data deduplication and compression. Its aim is to speed access for virtual server and desktop infrastructures with an architecture that comprises high-performance RAM drives as a front-end buffer that retain the hottest data in front of 1 TB MLC flash drives.

The device’s capacity ranges from 3.5 TB (15 TB effective) to 13.5 TB (60 TB effective), and it can deliver 120,000 4K IOPS. The Solidarity has four 1 GbE ports and two 10 GbE ports.

Greenbytes claims a better cost per gigabyte of storage than traditional arrays that use 15,000 rpm Fibre Channel drives. With deduplication and hardware compression, the claimed cost benefit is 10x and 150x in performance.

As befits an SMB-targetted product, the Solidarity has a wizard-based interface to provision iSCSI LUNs and file systems and carry out monitoring.

The Solidarity is one of a new breed of storage devices aimed specifically at virtual server implementations, such as those from Nutanix, Tintri, NexGen, Starboard and Tegile.

What bothers me at the moment is that this must be a very niche market. Virtual servers are simply not ubiquitous at the moment. The V-index survey carried out in the last quarter of 2011, for example, has the number of servers virtualised at around 35% in the UK. That’s among firms of more than 1,000 employees, so not really SMBs, but don’t SMBs lag further behind in virtualisation than bigger outfits?

The number of all-virtualised organisations must be quite small. And for those that are part virtualised, it must be a less-than-optimum choice to add dedicated virtual machine storage to an existing infrastructure. It’s always going to be cheaper and easier to add SSD to existing servers or arrays, surely?

Or if you want to add an SSD appliance you’d want it to act as a distinct tier within the existing infrastructure, a high-performance level added to lower-cost back-end spinning disk. It would certainly seem to make financial sense to shunt less-used data off to 7,200 rpm SATA instead of it squatting on MLC flash.

I asked Greenbytes whether the Solidarity could be used as a tier with other arrays. The answer was no, other than by use of storage virtualisation products, as is always the case when tying together storage arrays from different vendors.

Greenbytes’ EMEA vice president, Paul Silver, is confident the company is selling into a specific market demand, namely the SME, possibly going for its first SAN to support virtualisation, particularly of desktops.

He said, “The point is to offer a fully solid-state array with the capacity and price to match a SAS array and outperform it by a long way. We’re not trying to say one size fits all, but we’re trying to provide a breakthrough product with more performance than anyone else can offer at the price.”

Well, the proof will be in the pudding, as they say. Greenbytes has no customers for Solidarity right now. It’ll be interesting to see how it fares in the coming months.

And talking of price, you can get 3.5 TB for £37,000 and 13.5 TB for £75,000.

Follow me on Twitter: AntonyAdshead

It’s an idea that has some overlaps/parallels in other developments in storage, where numerous ideas aimed at lessening the distance between processing and media are emerging. Nutanix, for example, sells combined server and storage boxes that can be clustered grid-style. Indeed, EMC itself has its Project Lightning, which adds server-side SSD cache, a la Fusion-io.

EMC’s project with no name aims at the same goal — cutting the distance between storage and processing — but by having storage absorb the processing function. It could work, but by the same token there are numerous questions posed. Would the app have dedicated access to the controller processing power or risk contention from other hosts? Or to sidestep this, would the EMC box become like a giant DAS?

There are many more questions than answers right now.

Follow me on Twitter: AntonyAdshead

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.