In Fibre Channel's walled garden not a lot -technically - has been happening. A lot has been going on. For example, the 2 to 4Gbit/s product transition is all but over. EMC's InVista is still not in action. Cisco is is still not seeing any prospect of earning $1billion a year from storage networking and consolidation is proceeding apace with Brocade acquiring McData.
But technically the most noteworthy thing is McData's use of 10Gbit/s inter-switch links.
Non-technically the most noteworthy thing to me is Brocade's attempt to break out of the Fibre Channel walled garden with its broadened Tapestry software program. If we couple this with EMC's headlong rush into non-storage hardware acquisitions we might begin to re-arrange the pieces of the storage market in a new way.
Why has EMC bought VMware, Documentum, RSA, Legato and Dantz? Why hasn't EMC answered the nimble technical advances made by, amongst others, 3Par, Xiotech, Pillar Data, Acopia and Isilon?
The supposed answer has to be that it sees more money to be made in the long run from what it is calling IIM - Intelligent Information Management - than from selling periodically technically refreshed storage hardware. The last really impressive piece of storage hardware it brought out was Centera and that was in 2004. We've seen the industry produce SATA and SAS, FATA an perpendicular recording, the Blu-ray and HD-DVD fiasco, continuous data protection and virtual tape drives but what has actually happened to excite us in the Fibre Channel storage world?
The use of iSCSI as a Fibre Channel SAN extension doesn't seem to be happening. The promise of virtualisation is getting bogged down in information lifecycle management (ILM) where the need to have logically separate silos of disk storage - the different tiers - conflicts with disk storage virtualisation's idea of a single logical pool of storage. ILM itself is getting stymied by the need for data classification and the unclear benefits and methods of its adoption.
What is going on? Are we seeing a kind of technology paralysis in the Fibre Channel world?
Yes, there is the debate about where to have the disc virtualisation and layered storage management intelligence? Should it be in a clever director as IBM and EMC and Cisco would have us believe? Or should it be in a highly-intelligent disk array front end as HDS says it should?
I'd like to ask whether this question matters much at all? Will we actually ever see transparent ILM with data files being moved by automated policies from an EMC Clariion to an HP EVA array and logical maps being automatically updated so accessing servers don't know or care what's happened?
Elsewhere there has been a sustained burst of creative energy: NAS has become clustered and high performance with Isilon, Acopia, NetApp and others; storage arrays have become modular and much cleverer, witness products from 3Par and Pillar, as well as cheap and capacious like the Nexsan products. Continuous data protection has taken off with even Microsoft getting in on the act but in a somehow typically annoying half-finished kind of way. Data product management backup recording with Bocada and Crosswalk has arrived.
Practical SMI-s products are arriving courtesy of AppIQ and the associated Aperi competition.
Compliance regulations are forming an unholy alliance with e-mail archiving products such that good data retention, search, and discovery products are coming.
Potentially mouth-watering de-duping technologies have arrived from companies like Diligent with a claimed 25:1 compression ratio. There is a debate about whether this is compression or de-duplication. Who cares? It's a sterile question.
A 25:1 compression ratio gives virtual tape systems an enormous fillip in the job they have to do.
Encryption products have come to the fore with companies like Decru and DISUK.
iSCSI and IP storage is taking off with EqualLogic, Left Hand Networks and Zetera.
Even Linux is getting in on the storage act with that radical Clever Space concept of distributing RAID discs and parity data between storage locations.
There is even a potential return to the virtues of direct-attached storage with revolutionary Sun products such as Thumper and Honeycomb.
So much is going on in storage yet the main strategy of companies like EMC and Brocade is to get out of an earnings-limited area which is how they appear to characterise storage hardware and software.
Tell that to NetApp which has just become the terrabytes shipped leader in networked storage according to IDC. Tell it to Isilon, Acopia, Bocada, 3Par, Crosswalk, Pillar Data and the other emerging and successful storage technology companies.
Is the Fibre Channel SAN world becoming like the mainframe channel world from which, in part, it sprang? It will always be there, a fact of life for larger enterprises, much like, possibly exactly like, how mainframes are, but of little or no importance whatsoever to the rest of us?
I don't know. I just pose the question? Why is the Fibre Channel SAN area looking as if it is suffering from a severe dose of technology paralysis whilst the rest of the storage market is bursting with restless and creative energy?