Holy Rabbit Batman! Take that merge factor down. 100,000 is much to high. I think I have seen that above 90 wont help you much. 100,000 is just insane though...try like 50 and call me in the morning.
- Mark On 1/8/07, Ali Salehi <[EMAIL PROTECTED]> wrote:
Hello, I have a question about the scalability of the Lucene. I'm a lucene beginner and I would like to use it to index several million measurements (400 Millions). A measurement has a type, owner, id, precision and data. As an experiment, I tried to insert 5M values into a lucene index using compound index with merge factor 100,000. For searching I have two problems : 1. The search time for simple queries such as precision:\+0002 is really high (4-10 seconds). I want to know if this search time is normal considering the amount of data I inserted to the lucene (5 Million values)? If not, how can I improve it. I'm sure I can improve it by upgrading my current box (1G memory and 3.2 Ghz CPU with 2 MB cache). I'm looking for software/configuration solutions ? 2. The search gives TooManyClauses exception when I'm searching for a data item with the queries similar to the one below : precision:\+0002 AND data:\+0.85* I guess this a bug ?! Thanks for your help, Ali Salehi ************************************************************** Ali Salehi, LSIR - Distributed Information Systems Laboratory EPFL-IC-IIF-LSIR, Bâtiment BC, Station 14, CH-1015 Lausanne, Switzerland. http://lsirwww.epfl.ch/ email: [EMAIL PROTECTED] Tel: +41-21-6936656 Fax: +41-21-6938115 --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
