At some point, every dutiful analyst is asked to make a prediction about the future of storage. Well, here we are....
Since things don't happen in our industry in anything but a smooth fashion -- in other words, new technologies take time to be accepted and old technologies don't just die off in a year -- let's review what took place in 2005 and then project into the future.
Similarly, archiving took hold in 2005, especially for e-mail. I must say that many people in IT are still confused about the difference between backup/restore and archiving, but more and more folks seem to get it and have started installing archiving systems for their most mission critical data. Just look at the sales of EMC Centera and NetApp's NearStore and it's pretty obvious that IT installed these by the boatload in 2005. Specialty data protection/archiving products from companies such as Mimosa and Storactive should also do well for the same reason in 2006.
In the grand scheme of things, data protection overhaul is still in its infancy. I expect this trend to catch fire in 2006. 2005 was the year of narrowing the choices and phasing the chosen systems into production. This year will bring mass installations of disk-based backup and restore and archival systems throughout the enterprise. Whether driven by compliance or not, I expect companies throughout the world to be busily buying and implementing these products in 2006.
I would make a very similar statement for branch office (or remote office) overhaul, except it is one or more years behind the disk-based data protection trend described above. There is no question in my mind that IT now generally understands that data in remote locations is just as important to manage as data in the datacenter. Branch office consolidation from an IT perspective did take hold in 2005, but 2006 should see the bend of the curve. I expect products from companies like Tacit, Cisco (Actona) and Riverbed, and even players coming in from a WAN optimization or application acceleration space (Packeteer, Orbital, SWAN labs and the like), to catch the wind in 2006. The US financial companies led the pack in testing these products in 2005 and demonstrated that the trend is real. Now everyone wants in.
Expect iSCSI to finally make it into the datacenter this year, even if most applications it would run would still be tier 2. The total iSCSI numbers should at least double. Ten-gig iSCSI systems will debut but without much impact in terms of installations. They will present themselves as a true alternative to FC and IT would demand real benchmarks. Not much more than this will happen in 2006.
CDP showed up as real products in 2005, but the number of installations could be counted on the fingers of two hands. Expect CDP to go mainstream in 2006, especially with HP and EMC driving sales. However, I don't expect the total sales to be but a tiny fraction of the market. Companies focused on SMB markets, e.g. InMage, should do very well. Microsoft, with its near-CDP product DPM, and Symantec with Panther, should have their first real sales but the numbers will still not be big. 2007 will be the year for CDP.
Enterprise-level virtualization products will barely make a dent in 2006, but the engines will start. SMB virtualization products have taken a good hold already, with products from FalconStor, DataCore and StoreAge doing reasonably well in 2005. That should continue, but enterprise-level products, such as EMC Invista, will undergo serious evaluations. However, not many of these products will go into production. The industry will settle in on the SPAID method of virtualization for the enterprise.
I do not expect any real strides in the storage management area. Aperi will flounder and will be finally given to SNIA to manage. EMC and others will continue delivering proprietary products (ECC) but SMI-S interfaces will become commonplace. In other words, new hardware will come with SMI-S providers so that others can manage it. However, I do not expect much in terms of true large scale implementations of SMI-S.
Expect storage security to be finally accepted as a true "budgeted" item in 2006 across most industries. Thus far, security made it above the line only in a few industries (financial and government, for instance).
SAS will fight for its space, and I think it will find a foothold in the midrange. Remote replication using network-based products will be accepted as a genuine alternative to array-based replication and go mainstream.
Surprising as some would find it coming from me, I think Sun will improve its position in the storage industry by at least a few points. HP should do the same. They both showed signs of stabilizing in 2005, and I expect that will payoff in 2006.
Of all the companies, I think NetApp has the most interesting year ahead. It will finally roll out the Spinnaker-based product line. How well the transition occurs will determine their revenue and profitability. Two thousand and six is probably the most important year for NetApp in a decade. I bet EMC can't wait for a mess up.
All in all, I am very bullish on 2006 from a storage perspective. It has been a long time since we have seen increases in storage sales almost across the board.
But storage isn't the only thing I care about. I also predict that my golf game will still stink at the end of 2006.
Do you know…
About the author: Arun Taneja is the founder and consulting analyst for the Taneja Group. Taneja writes columns and answers questions about data management and related topics.