willholley commented on issue #2129: _find performance issues
URL: https://github.com/apache/couchdb/issues/2129#issuecomment-523897506
 
 
   One difference between Query and MapReduce is that Query will always load
   the entire document for each result row (effectively include_docs=true in
   MapReduce terms) and then construct a result returning only the specified
   fields. This means there's a potentially expensive additional fetch for
   each row, particularly so if your documents are large.
   
   In general, indexing with Query is significantly faster than JavaScript
   MapReduce but querying is not (though is more flexible).
   
   On Thu, 22 Aug 2019 at 13:30, Nicu Spinu <[email protected]> wrote:
   
   > Even if I create the index on runtime, or if using one that was created
   > before - the result will always take over 3.5seconds.
   >
   > The above piece of code is just to get some runtime outputs(they both run
   > ONE time):
   >
   >    1. The _find - using an index - takes 3700ms.
   >    2. Executing a view(that was created before) - 800ms
   >
   > They both examine the same amount of data(i.e 15846) - however - the _find
   > is way slower.
   >
   > —
   > You are receiving this because you are subscribed to this thread.
   > Reply to this email directly, view it on GitHub
   > 
<https://github.com/apache/couchdb/issues/2129?email_source=notifications&email_token=AAAX36ZJVJGEM6YQB44TNXTQF2BFBA5CNFSM4IOSIVN2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD445Q6Y#issuecomment-523884667>,
   > or mute the thread
   > 
<https://github.com/notifications/unsubscribe-auth/AAAX364Q7YBTSXNMHJE6TKDQF2BFBANCNFSM4IOSIVNQ>
   > .
   >
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to