A question about rendering data on server side VS client side.

DanDong

Hi everyone,

I have been thinking about this question recently and still have some parts that I feel confused. I hope to get more ideas. When I program the code in my projects, I don't give too many thoughts on such question. I think it is time to get some clear and general ideas on this.

So typically what is the benefit of rendering data on client side and what is the benefit of  rendering data on server side?

We usually first need to publish a map service in ArcGIS Server and then add either dynamicmapservice or titlemapservice in xaml page or cs page referencing the url of the map service. This process I think we are rendering the data on the server side and then use the data. Do we rendering the data to the client in this process? If not then do we request to the server when zoom in/out the map? I feel very confused on this part...I think we don't make such request to the server (in web adf, all the map action make request to the server), otherwise it will slow down the response speed.

If we perform a query task and display the results on a graphcislayer, we need to add a graphicslayer in xmal or cs page first. So adding a graphicslayer should be rendering data on the client side since there is no service url referenced for the graphicslayer.

Then how about featurelayer? It refers to a particular layer in the service so it actually has a url reference, which means featurelayer is rendering data on the server side? Or it just make request to the server, retrieve graph feature data and render the data on the client side?

DominiqueBroux

We usually first need to publish a map service in ArcGIS Server and then add either dynamicmapservice or titlemapservice in xaml page or cs page referencing the url of the map service. This process I think we are rendering the data on the server side and then use the data. Do we rendering the data to the client in this process? If not then do we request to the server when zoom in/out the map? I feel very confused on this part...

For dynamic mapservice or tiledmapservice, the rendering is done on the server side. Only images are transferred from the server to the client.
When zoom in or zoom out, the image(s) are requested to the server (except if the image is already available in the browser cache).

So adding a graphicslayer should be rendering data on the client side since there is no service url referenced for the graphicslayer.
Then how about featurelayer?

Right and a feature layer is just a kind of graphicslayer. the rendering is done on the client side.

Or it just make request to the server, retrieve graph feature data and render the data on the client side?

That's the right answer.