Lucene attempts to drive the query by the clause that's least frequent, so order or your clauses will not matter.
But, it uses a simplistic heuristic to do so: it looks at the first docID for each sub-clause and then reorders them in decreasing docID order. This isn't a perfect optimization since conceivably a low-count sub-clause could just happen to have eg docID 0 matching. Mike On Tue, Apr 20, 2010 at 6:12 PM, John4982 <john4...@gmail.com> wrote: > > Hi > > does Lucene search uses short-circuit when i execute query like: > > A:10 AND b:20 AND c:30 > > In general, does position of field names can impact search performance e.g. > if field A with value 10 is more frequent is this mean that this will be > slower than if value 10 is less frequent? > > best > John > -- > View this message in context: > http://n3.nabble.com/Short-circuit-in-query-tp738551p738551.html > Sent from the Lucene - Java Users mailing list archive at Nabble.com. > > --------------------------------------------------------------------- > To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org > For additional commands, e-mail: java-user-h...@lucene.apache.org > > --------------------------------------------------------------------- To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org For additional commands, e-mail: java-user-h...@lucene.apache.org
