We saw today in the OneData presentation that they try to unify access to file storage backends. This has some analogue in the concept of the "logical file catalogue" of the legacy "grid" days... more specifically in the fact that there was a common access using higher-level transport protocols.
This reminded me of the way that SAGA abstracts and standardises access to different middleware interfaces. Can someone comment on the analogue between these ways of integrating resource providers ? What does OneData do that SAGA implementations or specific middleware stacks can't do ?