2-3G index data with 20M records, and each record has 10 or more fields, 3-5 fields are group fields, which I want to count them for returned results after searching. These group fields contains about 1000 or more values. It means that I would count 1000 categories for each search.
Due to the searched results number will be very big, so iterator each record with HitCollector that is too slow to count all of records (counting will cost about 30 seconds) Unfortunately, If I use QueryFilter to query each category one by one and save it to BitSet, but the number of category is about 1000, I've to filter it about 1000 times to count each category. It still consumed about 20 seconds. Is there any better way to count these categories faster and low memory? Any idea is appreciated! -- View this message in context: http://www.nabble.com/How-to-count-categories--tf4500720.html#a12835751 Sent from the Lucene - Java Users mailing list archive at Nabble.com. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
