Sun's StorEdge 6920 array can now virtualise EMC CX (CLariion) and HP EVA arrays alongside Sun's own drive arrays. The enhanced feature set includes storage pooling, data replication, mirroring, snapshotting and ILM-type date migration across storage tiers, all across multi-vendor arrays.

Sun wants to virtualise the data centre into a single flexible pool of resources. With the enhancements the disparate arrays are seen and managed through a single user interface. Sub-pools of the storage can be dedicated to specific applications. In this the system parallels Solaris 10 containers. (See the note below for Sun CEO Jonathan Schwarz's blog on this point.)

Up until now the HDS approach to storage services was unique, odd even. Greatly enhance a disk controller and virtualise every other vendor's arrays. Then add storage services on top of the virtualisation. No other vendor did this. They have a specialised storage server taking care of this stuff with client software on all the servers acessing the SAN that this box would control access to. Or there are specialised boxes in the SAN fabric that do it.

We covered this aspect of storage service location in a previous SPAID article and also here and here

Now Sun has folowed the HDS route and greatly enhanced its 6920 controller. It's no longer a disk controller. Think of it as a storage utility processor. It's generally equivalent to TagmaStore in terms of features. And of course, SUN OEMS the TagmaStore system. James Whitemore, a Sun VP Marketing, says that it's no surprise the two virtualising systems look the same; Sun and HDS have: "a deep relationship." But the 6920 is: "a mid-range system in terms of price; it's mid-range in capacity terms, its features and functions are equivalent to TagmaStore's."

Pirus comes good
The new 6920 features are based on technology Sun acquired by buying Pirus two years ago. Whitemore says: "It is a fulfillment of a lot of the Pirus acquisition; a little over two years. It's great to actually get it to market."

Unlike TagmaStore which is focused at the enterprise level, the 6920 is determinedly mid-range. Whitemore says: "In the mid-range storage market the key thing is consolidation." With the 6920, Whitemore says, "Its consolidation features and functions are top of the priority list. We take complexity out of the storage environment by enabling a single management interface. No other vendor has this type of functionality in the mid-range. This is the first one."

Customers can consolidate heterogeneous arrays under a single interface. The current set of supported arrays will increase. HDS mid-range arrays? "Yes, absolutely." IBM TotalStorage? "Exactly, exactly."

Can the 6920 have multiple types of disk? Not in the 6920's own cabinet; that's Fibre Channel disk. But there can other types of disk, such as SATA drives, in arrays underneath the 6920. So we have the basic storage tiers for ILM. The announced crop of 6920 enhancements are probably only the start.

Location, location, location
It strikes us at Techworld that, if you buy a TagmaStore or 6920 system and get your virtualisation-based storage services the storage side of the fabric, then you are unlikely to buy a fabric-based or server-based storage services product. As (real) estate agents say, it's location, location, and location.

The message here is that Sun and HDS see no merit (to them at least) in an intelligent storage fabric. The (storage fabric) network is not the computer, so to speak. Sure, in a general sense, the storage network is the storage computer, but the processing power is not located in the network (fabric boxes). They just run data traffic management functions.

The Enterprise Strategy Group supports the Sun approach. Senior analyst Tony Asaro says: "The Sun StorEdge 6920 system is an innovative and compelling storage solution that should impress even skeptics." Especially if they are skeptic about IBM's storage tank, Cisco's MDS9000 storage services platform and EMC's Storage Router.

When will it be available? From the first of July this year.

Sun President and COO Jonathan Schwarze's blog on containers:-
"One of the key features in Solaris 10 is just this - "containers" are logical partitions that allow a single computer to behave like an unlimited number of smaller systems, with little/no overhead. Reboot a partition in 3 seconds, keep disparate system stacks on the same computer, assign different IP addresses or passwords to each, treat them like different computers, and use them to consolidate all those otherwise 15 percent utilized machines - sky's the limit (on any qualified hardware platform), and with it, customers can now drive utilization through the roof. With no new licensing charges. (And personally, I'm a fan of 3 second reboots.)"

"... Simultaneously, much the same inefficiencies described above have been plaguing the storage world. A few years back, "SSP's," or storage service providers, began aggregating storage requirements across very large customer sets, providing storage as a service. Most SSP's found themselves stymied by the diversity of customer they were serving. Each customer, or application opportunity, posed differing performance requirements (speed vs. replication/redundancy vs. density, eg). This blew their utilization metrics. Before the advent of virtualization, SSP's had to configure one storage system per customer. And that's one of the reasons they failed - low utilization drove high fixed costs.

"So that was the primary motivation behind the introduction of containers into our storage systems. The single biggest innovation in our 6920's is their ability to be divvied up into a herd of logical micro-systems, allowing many customers or application requirements to be aggregated onto one box, with each container presenting its own optimized settings/configurations. This drives consolidation and utilization - and when linked to Solaris, allows for each Solaris container to leverage a dedicated storage container. Again, driving not simply scale, but economy."