Can I cache my data for TimeSlider tool?

[JoshV]

I'm wondering if the TimeSlider will work on data that is cached?  Reason being, that I'm getting some inconsistent results on my Silverlight application that uses a TimeSlider.  I'm pretty much using the exact sample that ESRI has here (http://help.arcgis.com/en/webapi/silverlight/samples/start.htm#TimeMapService) but the Dynamic Service I'm using has two featureclasses that have over 10,000 features and my Silverlight Application doesn't always show all the data in TimeSlider mode Dynamically creating the cache

[ChristopherHill]

the TimeSlider is just used to pick a StartDate and EndDate which we refer to in the API as a TimeExtent. The TimeSlider is usually bound to the Map.TimeExtent property, when the user interacts with the TimeSlider it changes the TimeExtent property on the Map. when the TimeExtent property of the map changes it loops through its layer collection looking for layer types that support time data. If it finds a layer type that is TimeAware then it will request the data from the service and pass along the time values so that the returning data matches the TimeExtent. There could be caching in two places, caching can be occuring on the REST service, when the REST service see url that is identical to a request that it recently repsonded to, then it will pull it from cache to be more preforment, The second place you have caching is in the browser, if the browser sees a url that is identical to one it has already responded to it will return the cached response.

One more thing to note is there is a limit on how many features will return from the server, I believe the limit is 1,000 features for a single service request for ArcGIS Server v10. It was only 500 features in v9.3. So if there are 10,000 features your only getting 1,000 of those features. So If you are using the TimeSlider and requesting a small date range you may get back 500 features, but then you increase the range which is valid for 1,500 features, but only 1,000 features will come back. Hope this information helps.

[JoshV]

Great explanation Chris, I appreciate it.  How can I increase the features to something higher than a 1,000 features?

[dotMorten_esri]

This is a setting in an config file in the REST WebService.
However I would generally recommend against this. It's a LOT of features to bring down on the client and is not going to perform well (slow startup-time and poor rendering performance). Instead render your features using ArcGISDynamicMapServiceLayer instead of FeatureLayer, which is specifically optimized to render map objects very fast.

[JoshV]

This is a setting in an config file in the REST WebService.
However I would generally recommend against this. It's a LOT of features to bring down on the client and is not going to perform well (slow startup-time and poor rendering performance). Instead render your features using ArcGISDynamicMapServiceLayer instead of FeatureLayer, which is specifically optimized to render map objects very fast.

Hi Morten-  Is this setting you are referring to called UserServiceCacheLimit??  Is the file you are referring to in inetpub\wwwroot\ArcGIS\rest\rest.config? Also if I make a change to that file do I need to restart the SOM to see changes?

many thanks

[TerryGiles]

Josh,

You can set the maximum number of features returned either in ArcCatalog (with an Admin connection to the server - version 10 only), Server Manager (on the Parameters tab when viewing a Service's properties) or by editing the config file for the service, which is typically found in c:\program files\ArcGIS\Server10.0\server\user\cfg.  In the .cfg file for the service, set the value for the <MaxRecordCount> tag.  I believe you have to restart the SOM for this to take effect.