Heya Ray, On Jun 12, 2008, at 02:51, Ray Hilton wrote:
CouchDB sounds like the holy grail for Large storage... but what happens if you hit the disk limit of a single machine? Is there a way to host parts of the index on different machines (in a redundant and fault tolerant manner)? I was thinking of writing my own that would partition by date and then inspect the query to see which indices to include, and send a MapReduce task to whichever machines had a copy of the relevant indices. Is this something that may be possible with CouchDB?
CouchDB is not there yet and we don't know exactly how things will look, but this is very much on our roadmap. Cheers Jan --
