As David suggests, the reason storage options are memory and file are for simplicity and correctness. Performance and scalability were not requirements, because they are not (yet) requirements for Wave in a Box. The file based delta store demonstrates what is required of any delta store implementation with well-understood semantics.
I'm not aware of a concrete roadmap. I personally would like to see an SQL-based backing because of the overwhelmingly wide experience with and use of SQL databases in the technology world. Delta persistence makes little to no use of relational constraints so I think this would make for a robust and scalable solution. There is also a lot of community demand for one of the many NoSQL databases, but I don't believe there's any significant technical advantage to be had there at the moment. A. On 22 March 2011 12:13, Jeremy LeBard <[email protected]> wrote: > Hi All, > > I was wondering about the architecture for persistent storage options and > Deltas. Is there a reason why the storage is memory, file and not MongoDB? > Is file best in terms of performance and scalability? Whats the roadmap for > the future? > > Also, the Delta storage option says "not stable and shouldn't be relied > upon > for long-term storage...". What is the time-frame or goal for this option > to > mature? > > Thanks! > > Jeremy >
