Question

After reading http://msdn.microsoft.com/en-us/library/system.servicemodel.description.servicethrottlingbehavior.maxconcurrentsessions.aspx

and

http://msdn.microsoft.com/en-us/library/system.servicemodel.description.servicethrottlingbehavior.maxconcurrentcalls.aspx

I have concluded that:

MaxConcurrentSessions is the number of queued sessions per client (default of 10)
MaxConcurrentCalls is the number of active connections on the service (default of 16) i.e. all clients accessing the service at any one time, meaning that if 2 client did 10 calls each, 4 would have to wait in the queue for processing.

Questions:

1. Is my conclusion correct?
2. How does MaxConnections interact with these?
3. Does MaxConnections take precedence over the MaxConcurrentX settings?

(Note:I am using .NET 3.5)

Answer 1

MaxConcurrentCalls has to do with the number of calls on the service that are currently executing.

MaxConnections has to do with the total number of open connections on the service, regardless if the service is executing anything for the connection.

For example, if a client opens a connection to the service, calls a method, and is waiting for the method to return, it will count against the MaxConcurrentCalls. As soon as the service returns a response to the client’s method call, it will not count against the MaxConcurrentCalls… even if you didn’t close the client-side proxy. Assuming you didn’t close the client-side proxy, the connection would count towards the MaxConnections on the service since you still have the connection open, but it’s not currently executing anything on the service so it would not count against the MaxConcurrentCalls.