Object Storage moving forward?
Object Storage has the potential for being the engine for storage's expected explosive growth. It scales, and as the Gridstone team points out, it can run on inexpensive hardware. The issue has been weak performance. This is the result of a focus on scalability as the way to get both capacity and performance, rather than node performance, per se. If Object Store performance approaches traditional Big Iron storage, while enjoying 'PC' costs, we are likely to see a rerun of the mainframe-to-server evolution in the storage market.
Gridstone may be a stepping stone (no pun?) to getting more out of each node. I await benchmark publication with my fingers crossed!
Biting the hand that feeds IT
