Andy, I opened this ticket so that someone can eventaully invistigate: https://issues.apache.org/jira/browse/SOLR-4874
Just an instanity check, I see I had misspelled "maxCollations" as "maxCollation" in my prior response. When you tested with this set the same as "maxCollationTries", did you correct my spelling? The thought is that by requiring it to return this many collations back, you are guaranteed to make it try the maximum time every time,giving yourself a cleaner test. I am trying to isolate here if spellcheck is not running the queries properly or if the queries just naturally take that long to run over and over again. James Dyer Ingram Content Group (615) 213-4311 -----Original Message----- From: Andy Lester [mailto:a...@petdance.com] Sent: Tuesday, May 28, 2013 4:22 PM To: solr-user@lucene.apache.org Subject: Re: Why do FQs make my spelling suggestions so slow? Thanks for looking at this. > What are the QTimes for the 0fq,1fq,2fq,4fq & 4fq cases with spellcheck > entirely turned off? Is it about (or a little more than) half the total when > maxCollationTries=1 ? With spellcheck off I get 8ms for 4fq query. > Also, with the varying # of fq's, how many collation tries does it take to > get 10 collations? I don't know. How can I tell? > Possibly, a better way to test this is to set maxCollations = > maxCollationTries. The reason is that it quits "trying" once it finds > "maxCollations", so if with 0fq's, lots of combinations can generate hits and > it doesn't need to try very many to get to 10. But with more fq's, fewer > collations will pan out so now it is trying more up to 100 before (if ever) > it gets to 10. It does just fine doing 100 collations so long as there are no FQs. It seems to me that the FQs are taking an inordinate amount of extra time. 100 collations in (roughly) the same amount of time as a single collation, so long as there are no FQs. Why are the FQs such a drag on the collation process? > (I'm assuming you have all non-search components like faceting turned off). Yes, definitely. > So say with 2fq's it takes 10ms for the query to complete with spellcheck > off, and 20ms with "maxCollation = maxCollationTries = 1", then it will take > about 110ms with "maxCollation = maxCollationTries = 10". I can do maxCollation = maxCollationTries = 100 and it comes back in 14ms, so long as I have FQs off. Add a single FQ and it becomes 13499ms. I can do maxCollation = maxCollationTries = 1000 and it comes back in 45ms, so long as I have FQs off. Add a single FQ and it becomes 62038ms. > But I think you're just setting maxCollationTries too high. You're asking it > to do too much work in trying teens of combinations. The results I get back with 100 tries are about twice as many as I get with 10 tries. That's a big difference to the user where it's trying to figure misspelled phrases. Andy -- Andy Lester => a...@petdance.com => www.petdance.com => AIM:petdance
