incubator-couchdb-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Brian Candler <B.Cand...@pobox.com>
Subject Re: Penalty for reduce?
Date Fri, 06 Feb 2009 07:44:37 GMT
On Thu, Feb 05, 2009 at 07:09:15PM -0800, Chris Anderson wrote:
> On Thu, Feb 5, 2009 at 12:38 PM, Brian Candler <B.Candler@pobox.com> wrote:
> > I'm just wondering, how much is the performance penalty for adding reduce to
> > a view? In other words, is there any reason *not* to include a reduce
> 
> I'm not sure. It would be a simple matter to add log() statements to a
> reduce function to see if it is run during map queries or if it is
> only run when reduce=true.

I was also wondering how much penalty there is when the view is being
updated. (I would expect not much - I understand the reduce value is stored
in each interior b-tree node, so would only have to be recalculated for
those nodes which have changed, but wanted a second opinion)

Mime
View raw message