msokolov commented on issue #701: LUCENE-8836 Optimize DocValues TermsDict to 
continue scanning from the last position when possible
URL: https://github.com/apache/lucene-solr/pull/701#issuecomment-500131574
 
 
   bq. The approach I took was to run some Lucene tests while counting the 
total number of seeks and terms read in the IndexInput, with and without the 
optimization.
   
   The numbers look compelling. Is this a fair test though? Doesn't it ignore 
the cost added by the optimization? We now do some work saving last term,etc. 
It seems as if it could be justified by reductions in seeking, but a test that 
shows that holistically would be good, ideally targeting some common use case. 
The unit tests might be exercising fairly artifical edge cases? EG we probably 
don't want to be optimizing slow exact range query when the caller would be 
better off using a points field. For DocValues, typical use cases do probably 
involve sorting/scoring/grouping/aggregations. Do you have a use case you are 
targeting and can share results on?

----------------------------------------------------------------
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

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to