Is there any formal definition for WMS Service Cascading?

I saw the term online, trying find any formal definition, thanks

Is there any formal definition for WMS Service Cascading?

I have implemented a couple of examples of cascading. It is very useful and powerful technique.

A good example is when departments or organisations are managing separate data but ultimately want to publish a single view.

Organisation A can host the raw data behind their own WMS server, for example they may host full world photography coverage as a single WMS layer "WorldPhoto" published as JPEG.

Organisation B can then reference the layer "WorldPhoto" as a layer in their own server.

Organisation B can customise "WorldPhoto" within their own WMS server. For example, switching off visibility over 1:1250, and adding a road network over the top, publishing the aggregated result in PNG format as another layer "OrgBMap".

It is a very powerful model, because the original raw data remains in the ownership of the correct department / organisation. Subsequent steps in the pipeline 'choose' to consume and republish that data, they do not have to maintain it.

creed's picture

Is there any formal definition for WMS Service Cascading?

Conceptually, WMS Cascading is a type distributed search. From the WMS 1.3 Implementation Specification: Cascaded layers
A Layer is said to have been "cascaded" if it was obtained from an originating server and then included in the service metadata of a different server. The second server may simply offer an additional access point for the Layer, or may add value by offering additional output formats or reprojection to other coordinate reference systems.

If a WMS cascades the content of another WMS then it shall increment by 1 the value of the cascaded attribute for the affected layers. If that attribute is missing from the originating server's service metadata, then the Cascading WMS shall insert the attribute and set it to 1.

Hope this helps!