Virtualization appliances do present a single, ubiquitous storage interface to servers on the front end and simplify the presentation of the different types and tiers of storage on the back end while allowing migration of data between the different tiers, but a clear understanding of the correct hardware for each unique infrastructure is still lacking.

This considerable gap between the promise and reality of a manageable tiered storage architecture means that users are often left to fend for themselves when deploying high-performance monolithic arrays alongside lower-cost midrange arrays that include a myriad of RAID configurations without a cohesive way to move data between the different arrays.

Part of the problem is that the lower costs of second and third tiers of storage lure users into buying the wrong tier of storage. Hitachi Data System's chief technology officer, Claus Mikkelson, finds that users who neither plan their storage strategy nor try to understand application requirements before procuring storage end up with a configuration that does not meet the application's needs. "As a result," Mikkelson says, "users end up in a trap from a performance and reliability perspective."

Another problem is that IT managers are often left alone to drive all the changes necessary to deliver on the benefits of a tiered storage infrastructure.

MidAmerica Bank IT manager Paul Stonchus manages an all-EMC environment that is made up of about 5 percent DMX, 65 percent Clariion and 35 percent Centera arrays as well as tape storage. But the Naperville, Ill., bank's mainframe computer output to laser disk (COLD) application is the only one that takes advantage of tiered storage.