Hi! Actually this happened because of copied viewserver design: CouchDB could pass documents to viewserver in async way, but this feature doesn't used(by no one afaik), so all documents have to be proceed one by one per each design document. Right now I'm trying to figure how to use this behavior for new python query server, but no numbers or results currently I have. But I'll inform you if I've got something, stay tuned(:
-- ,,,^..^,,, On Wed, Jan 4, 2012 at 5:42 PM, Yannick Schwartz <[email protected]> wrote: > Hi all, > > I'm relatively new to couchdb. I'm using couchdb-python for interacting and > building my views. > However, I expected my maps/reduces to run in parallel when launching a > view, but that doesn't seem to happen (according to my CPU usage). Did I > miss something or is that the expected behaviour? Or is this because of the > Python client? > > Thanks, > > Yannick
