"The user session states and aggregates occupy RAM above and beyond the RAM used to store the core unaggregated dataset. Most of the session information is shared between sessions in the same state" and how its done??
There is a central cache which stores all the aggregates per document( Aggregates means the aggregated data needed for the chart objects available in the qlikview document.)
So the RAM keeps all this information for each document -
1. Unaggregated Dataset( as per documents data model)
2. Aggregates( data needed for the chart objects, this is kept at central cache, As I know this gets updated as per server's memory management techniques as per usage of chart objects in the document. If a specific chart object is not opened/used/calculated by any user, I think it won't be existing here.)
3. User Session states(Each user needs to have his or her own User Session States, different users accessing same document are actually filtering, accessing document differently so user specific information needs to be maintained in different states although they might be sharing chart aggregates from central cache with different users.
