[Mapserver-dev] RE: postgis db connection caching...

Chris G. Nicholas cgn at globexplorer.com
Sat Jun 21 22:34:10 EDT 2003


should have figured as much :-)
I'll definitely look at the SDE adaptor; 

As far as mapscript goes, we're certainly not planning to fork()/exec() every click, but rather use the mapObj Java constructor on dynamically generated mapfiles, save the image, then call mapobj.delete() to free memory. Since the Tomcat process should stick around, perhaps a simple static list of connections might do the trick... 

Chris

-----Original Message-----
From:	Paul Ramsey [mailto:pramsey at refractions.net]
Sent:	Sat 6/21/2003 3:09 PM
To:	Chris G. Nicholas
Cc:	mapserver-dev at lists.gis.umn.edu
Subject:	Re: [Mapserver-dev] RE: postgis db connection caching...
Given thought to and implemented.
  Jan Hartmann committed a per-session caching scheme about 6 months 
ago. Layers which use the same connection can pull the handle from a 
connection pool.
  Read through the SDE layer to see it in action. Currently it is only 
used by the SDE layer type. We may add it to the PostGIS section, but 
frankly connection overhead is very small for PgSQL.
  As far as mapscript goes, the idea system would be one where handles 
could be passed into mapscript from an external application. But that 
could be really hard to do in a way which binds nicely with multiple 
languages.
Paul

Chris G. Nicholas wrote:
> Before I attempt shameless experimental hacking, has anyone given
> thought to caching db connections, both within the same mapfile
> across different layers, and in general within mapscript
> applications?
>

-- 
       __
      /
      | Paul Ramsey
      | Refractions Research
      | Email: pramsey at refractions.net
      | Phone: (250) 885-0632
      \_










More information about the mapserver-dev mailing list