Hi John, I heard of many users who used Lucene for this use-case, it's definitely a valid one. Indexes are stored mostly on disk, with a tiny part of them being held in memory to guarantee good access speed. Lucene supports both inverted indexes and KD trees up to 8 dimensions.
Lookup, sorting and pagination are all supported. Beware however that Lucene is designed to retrieve the top hits and paginating deeply is a usage pattern that Lucene won't shine at. Making changes visible to a Lucene index is a costly operation, which should be run on a fixed interval rather than after every document. We recommend to not make every document visible immediately. However, it is fine to make changes visible every second for instance. For the record, you might want to look into Apache Solr or Elasticsearch for this kind of usage, they would make things simpler. On Tue, May 28, 2019 at 2:18 AM John Dale <jcdw...@gmail.com> wrote: > > Greetings; > > I'd like to play around with Lucene to offload some of my database lookups. > > Is this a valid use of Lucene in your opinion(s)? > > Indexes - they are stored on the file system as some kind of tree (I'm > guessing)? > > Lookups and sorting - Can I lookup by date and sort asc/desc and paginate? > > Is Lucene transactional when adding to the index? > > Sincerely, > > John > > --------------------------------------------------------------------- > To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org > For additional commands, e-mail: java-user-h...@lucene.apache.org > -- Adrien --------------------------------------------------------------------- To unsubscribe, e-mail: java-user-unsubscr...@lucene.apache.org For additional commands, e-mail: java-user-h...@lucene.apache.org
