Also, what's the performance impact of range queries vs. querying for a particular DAY (as described in my last post) when the index contains , say, 10 million docs ?
If the range queries result in a significant performance hit, one option for us would be to define additional DAY fields when indexing TIME data eg. when indexing METADATA.DATE_ABC= 2009-07-31T15:25:45Z , also create and index something like METADATA.DATE_DAY_ABC= 2009-07-31T00:005:00Z -- View this message in context: http://lucene.472066.n3.nabble.com/Date-granularity-tp3920890p3924290.html Sent from the Solr - User mailing list archive at Nabble.com.
