NetApp Kills Off DataFort; Encryption A Feature, Not A Product

Reports that NetApp is killing off the DataFort encryption product line it acquired with Decru in 2005 signals the last gasp of the once-promising storage encryption market. While we argued over Data Domain and whether deduplication was a product or a feature, the market decided that encryption alone does not a product make.  

NetApp is talking up Brocade's encrypting switches, which include technology that Brocade licensed from NetApp/Decru as a replacement for DataFort. Decru customers that wanted to encrypt data at rest on their disk arrays could also get encryption add-ons from Cisco for MDS or use Emulex's encryption option for their HBAs. Frankly I'm not a believer that this kind of encryption at rest actually prevents much data loss, but if you are, you still have options.

Datafort was the last of what, for a while, appeared to be an industry of encryption appliances from deceased vendors like Kasten Chase and NeoScale (snapped up by nCipher, which was itself absorbed by French defense contractor Thales). While Thales continues to sell the Cryptostor tape encryptor, I don't understand why anyone would buy one.  

If you come to the reasonable decision that it's time to start encrypting tapes, you could get a Cryptostor and continue to use your old tape library. Alternatively, you could spend the same money replacing the drives in the library with LTO-5 drives that have built-in encryption and spend less on tapes in the future since the LTO-5s store 2 to 16 times as much data as my old drives. Heck, you might even save on maintenance.

The end of Decru's market life is also a commentary on NetApp's ability to manage acquisitions. The truth is, some organizations are better at acquiring technologies and making them their own than others.  Cisco's leading position in the Ethernet switching business came through the acquisition of switch pioneer Kalpana, (along with a few competitors), and while most storage guys think of Clariion as a pure EMC product, it was initially developed down the road at Data General.NetApp, on the other hand, has not managed to turn an acquisition into a successful product line. The company has had what can be called , at best, a modest success with Onaro's SANscreen SAN management tools and the NearStore VTL product line derived from Alacritus's technology. Topio's continuous data protection for heterogeneous environments became NetApp's ReplicatorX before the folks in the corner office decided to rename it, yet again, and kill it off.  

The jury's still out on the Bycast object storage tech that NetApp snapped up earlier this year, but I'm not that optimistic. The rumors I'm hearing are more about making Data OnTap and filers cloudier with Bycast tech than about NetApp really pushing a new object store product line a la EMC's Atmos.

Despite some well documented delays, NetApp has done better at integrating acquired technologies into Data OnTap with Orca, Troika and, most importantly, Spinaker tech, all of which make NetApp filers the Swiss army knives of storage they are today.