Doing a fast ID only search !

["Kraus, Ralf | pixelhouse GmbH" <rk...@pixelhouse.de>]

Hello,

I have a "little" performence problem with SOLR (again).
My searches delivering 30 rows per site (for my webpage) BUT I need to 
get about 500 primary keys (IDs).
Right now I am searching for my 30 complete rows (all fields) and 
another search with the setting "fl=primaryID".

Unfortunately the second (ID only) search take 3 times of my first (full 
search)....

Is there a chance I could do this a smarter way ?

Greets -Ralf-

Re: Doing a fast ID only search !

[Yonik Seeley <yo...@apache.org>]

On Mon, Nov 3, 2008 at 6:21 AM, Kraus, Ralf | pixelhouse GmbH
<rk...@pixelhouse.de> wrote:
> I have a "little" performence problem with SOLR (again).
> My searches delivering 30 rows per site (for my webpage) BUT I need to get
> about 500 primary keys (IDs).
> Right now I am searching for my 30 complete rows (all fields) and another
> search with the setting "fl=primaryID".
>
> Unfortunately the second (ID only) search take 3 times of my first (full
> search)....

Loading the stored fields (or even one) of a document is not a fast operation.
Do really need to get all 500 IDs back (i.e. what will you do with
those 500 IDs, and is there a different way of accomplishing the end
goal).

-Yonik