You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@couchdb.apache.org by Apache Wiki <wi...@apache.org> on 2011/11/23 16:15:38 UTC

[Couchdb Wiki] Update of "How_to_handle_stats_aggregation" by JanLehnardt

Dear Wiki user,

You have subscribed to a wiki page or wiki category on "Couchdb Wiki" for change notification.

The "How_to_handle_stats_aggregation" page has been changed by JanLehnardt:
http://wiki.apache.org/couchdb/How_to_handle_stats_aggregation?action=diff&rev1=2&rev2=3

Comment:
add ContributorGroup link

+ <<Include(EditTheWiki)>>
+ 
  I got an email today explaining an approach to stats logging that quite suprised me in terms of its complete obviousness (that I had failed to see) and its complete awesomeness.
  
  The basic idea: pre-reduce. The use case as described below is for stats reporting and the idea is that after we've figured out to use _bulk_docs and buffer some number or time period of log data before inserting, we can go one step further and pre-reduce that data down into a single doc for insertion. I can already see people implementing support via their favorite language binding support to even use the same map/reduce functions from CouchDB as a pre-processor for creating summary docs on input.