[Mapserver-users] Mapserver WFS performance

[Yewondwossen Assefa]

Here is how things work :
   - the user does a wfs request
   - in is translated in mapserver query : the query does not take into 
account the maxfeatures.
    - The MAXFEATURES is taken into account when generating the 
resulting   gml file

   So yes, It ends up doing the query on all the shapes in your case.

  I think we should modify mapserver low level query functions to be 
able to pass a maxfeature parameter. Some of the query functions have a 
mode (single or mulitple), and this notion should be extended to support 
a maxfeature. Once this is done the wfs interface can take advantage of it.

  Later,




Bart van den Eijnden wrote:

> Hi list,
> 
> Today I was testing Mapserver WFS on a shapefile of 450 Mb in size with 
> a quadtree index.
> 
> I am using no Filter (no BBOX) but only a MAXFEATURES parameter in my 
> URL. Perhaps this is not the intention of Mapserver WFS, but this is 
> also a potential for people to "attack" the server. It looks like 
> Mapserver is reading the whole shapefile in order to give back the first 
> 1 or 10 features. I stopped testing at 10 features because of the long 
> response times.
> 
> These are the results of my performance test:
> 
> MAXFEATURES    RESPONSE_TIME [seconds]
> 1              192
> 10             210
> 
> Any comments on this? Am I doing something wrong?
> 
> Best regards,
> Bart
> _______________________________________________
> Mapserver-users mailing list
> Mapserver-users at lists.gis.umn.edu
> http://lists.gis.umn.edu/mailman/listinfo/mapserver-users
> 


-- 
----------------------------------------------------------------
Assefa Yewondwossen
Software Analyst

Email: assefa at dmsolutions.ca
http://www.dmsolutions.ca/

Phone: (613) 565-5056 (ext 14)
Fax:   (613) 565-0925
----------------------------------------------------------------