If searches normally arrive via http, you could use any HTTP test client: 
http://www.hpl.hp.com/research/linux/httperf/ for example. If you normally send 
queries via XCC, you might look at 
https://github.com/marklogic/performance-meters

Think about factors that might influence performance. For example what queries 
to test, and how many clients to use. Or caching: when all data is in cache, 
there shouldn't be any difference between the two configurations.

What are you expecting to see? My general experience is that multiple forests 
have more impact on ingestion speed than query speed.

-- Mike

On 21 Aug 2013, at 08:37 , Christine Schwartz <[email protected]> 
wrote:

> Hi,
> 
> I recently created a new version for the database of our largest digital 
> library collection with two forests, rather than one.
> 
> Now, I'd like to compare the speed of these two databases--the old one with 
> one forest and the new one with two. I want to test search query speed.
> 
> What's the best approach to this kind of testing? Is it possible to do this 
> without bothering our developer to amend his code with query trace, etc.?
> 
> Thanks,
> 
> Chris
> 
> Christine Schwartz
> XML Database Administrator
> Princeton Theological Seminary Library
> [email protected]
> 
> _______________________________________________
> General mailing list
> [email protected]
> http://developer.marklogic.com/mailman/listinfo/general

Attachment: smime.p7s
Description: S/MIME cryptographic signature

_______________________________________________
General mailing list
[email protected]
http://developer.marklogic.com/mailman/listinfo/general

Reply via email to