First, unless you know that the Name field being on disk is a
bottleneck, start simple.  Put everything in a single index.

If you really need the "Name" part to be entirely in memory, then
create two indicies, one with all of your fields except Name and
another with Name only.  Then load the Name index into a RAMDirectory,
and create a ParallelReader over the two different indicies.

-Yonik
http://incubator.apache.org/solr Solr, the open-source Lucene search server


On 7/17/06, neils <[EMAIL PROTECTED]> wrote:

Hm,

so if following is possible than it would be great:

I have a table in SQL-Database with following columns: ID, Name, Forename,
Street, Phonenumber.

So i would make index for each column (ID will included in all indizies).

After this i have 4 indicies. Name (in memory), Forename (on Disk), Street
(on Disk), Phonenumber (on Disk).

When now somebody search for name "kos" in the street "teststreet" can you
use parallelreader for this and how ?

Thank you very much!
Neils

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to