You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-user@lucene.apache.org by nitinkhosla79 <ni...@gmail.com> on 2012/04/17 13:28:21 UTC
Merge results of search query to multiple cores
I have setup solr with multiple cores. Each core has its own schema(with
common unique id).
Example:
Core0: Id, name
Core1: Id, type.
I am looking for a way to generate the resultset as Id, name, type.
Is there any solution for such problems?
It is more like a database join query.
------------
Problem: I had to keep 2 cores. Because one index is going to very
large(indexed many pdf documents) and other is going to be very small but to
be very frequently updated(for every indexed document).
So combining the schema might make it impossible(with respect to time
needed) to index millions of documents.
------------
In previous posts(from 2010,) it has been mentioned that it is not possible
with solr. Wondering if this has changed.
This link below mentions on using 3rd core and then using shard to merge the
results. But for me, it does not merge but shows the results EITHER from
core0 OR from core1 for the SAME id which is present in both core0/core1.
http://stackoverflow.com/questions/2139030/search-multiple-solr-cores-and-return-one-result-set
Please suggest further.
--
View this message in context: http://lucene.472066.n3.nabble.com/Merge-results-of-search-query-to-multiple-cores-tp3916976p3916976.html
Sent from the Solr - User mailing list archive at Nabble.com.
Re: Merge results of search query to multiple cores
Posted by Erick Erickson <er...@gmail.com>.
Why do you have these in two cores? Why not
have them in a single core (two different
document types) and do a join?
Best
Erick
On Tue, Apr 17, 2012 at 7:28 AM, nitinkhosla79 <ni...@gmail.com> wrote:
> I have setup solr with multiple cores. Each core has its own schema(with
> common unique id).
> Example:
> Core0: Id, name
> Core1: Id, type.
>
> I am looking for a way to generate the resultset as Id, name, type.
> Is there any solution for such problems?
> It is more like a database join query.
> ------------
> Problem: I had to keep 2 cores. Because one index is going to very
> large(indexed many pdf documents) and other is going to be very small but to
> be very frequently updated(for every indexed document).
> So combining the schema might make it impossible(with respect to time
> needed) to index millions of documents.
> ------------
> In previous posts(from 2010,) it has been mentioned that it is not possible
> with solr. Wondering if this has changed.
> This link below mentions on using 3rd core and then using shard to merge the
> results. But for me, it does not merge but shows the results EITHER from
> core0 OR from core1 for the SAME id which is present in both core0/core1.
> http://stackoverflow.com/questions/2139030/search-multiple-solr-cores-and-return-one-result-set
>
> Please suggest further.
>
> --
> View this message in context: http://lucene.472066.n3.nabble.com/Merge-results-of-search-query-to-multiple-cores-tp3916976p3916976.html
> Sent from the Solr - User mailing list archive at Nabble.com.