[MapProxy] WMS vs WMTS ( speed )

Martin Kokeš shr3k at typo3-hosting.com
Sun Mar 3 03:52:40 PST 2013


I suppose there must be. WMS requests aren't (or shouldn't be) cacheable by browser unlike WMTS/TMS requests and MapProxy must these images stitch together from tile cache eventually cut it again to cilent side specified pseudo tiles or even worse, depending on cache/WMS configuration, reproject it or resample it to fullfill client's request. This is favorable in case of MapProxy acting as local WMS cache for busy public WMS service - you can seamlessly work with already seeded coverages with your favorable GIS or CAD client.

On the other hand in case of WMTS/TMS client must know and must exactly specify which tile he wants. Reprojection is possible only on client side (in example with help of OpenLayers/proj4js etc.). Besides in case of WMTS/TMS there's possibility to force nginx/apache to serve these seeded tiles directly from MapProxy cache with full speed you can achieve.


From: Marco Scheuble [mailto:marco.scheuble at gmx.de]
To: mapproxy at lists.osgeo.org
Sent: Fri, 01 Mar 2013 17:18:44 +0100
Subject: [MapProxy] WMS vs WMTS ( speed )

Hi All,
  I am wondering if there are differences in performance or speed when using wms or wmts. I think, when MapProxy generates the requested data out of the cache, there shouldn't be a difference? Or am I wrong with that?
  thanks in advance,
  MapProxy mailing list
  MapProxy at lists.osgeo.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/mapproxy/attachments/20130303/a58fd924/attachment.html>

More information about the MapProxy mailing list