The next thing we are going to cover is how the Connection Broker decides which of the RDS host servers in the farm that is running the requested application is actually going to deliver the application. There are two options for configuring load balancing.
For the first option, there is no real complicated science behind the load balancing from a Horizon View perspective. It is purely based on how many sessions are available on any given RDSH server. So, that means when a user logs in and launches a published application, the application is delivered from the server that has the highest amount of free sessions available; that is, the one that is least busy.
This is illustrated in the following diagram:
This first option works fine, but how does it know what each session is consuming in terms of resources? A particular host may well...