Indexed Shapefile vs. PostGIS

[Gambin Dejan]

Sorry, I posted it in a wrong thread, my intention was to reply on
"Benchmarking mapserver by simulating visitors" subject

sorry again

dejan

-----Original Message-----
From: UMN MapServer Users List [mailto:MAPSERVER-USERS at LISTS.UMN.EDU] On
Behalf Of Gambin Dejan
Sent: Monday, August 29, 2005 11:45 AM
To: MAPSERVER-USERS at LISTS.UMN.EDU
Subject: Re: [UMN_MAPSERVER-USERS] Indexed Shapefile vs. PostGIS


Thanks to everyone suggesting me the possible solutions.

Stephen, do you maybe know why siege on my debian machine gives me
"segmentation fault" whenever I run it (and whatever option I choose,
even -C)??

thanks very much

dejan

-----Original Message-----
From: UMN MapServer Users List [mailto:MAPSERVER-USERS at LISTS.UMN.EDU] On
Behalf Of Stephen Woodbridge
Sent: Thursday, August 25, 2005 1:53 AM
To: MAPSERVER-USERS at LISTS.UMN.EDU
Subject: Re: [UMN_MAPSERVER-USERS] Indexed Shapefile vs. PostGIS


Schuyler Erle wrote:
> * On 24-Aug-2005 at  1:46PM PDT, Ethan Alpert said:
> 
>>You can also use shptree to generate a spatial index file (QIX) which
>>mapserver will use to fairly efficiently display the data.
> 
> 
> I've found that the QIX support isn't sufficient to speed up access to

> really large shapefiles, especially if you're doing any kind of
> attribute filtering. I'd love to know why this is.
> 
> SDE
> 
Well, the QIX is a spatial index, not an attribute index. But someone 
might want to check the source code and see if attribute filtering makes

use of the subset of spatial objects or just dumbly scans all the 
attributes then rejects based on spatial criteria.

-Steve W.