Hi,

I have a Master/Slave application, which is using remoting (TCPChannel) for 
communications. Right now i'm testing a computational domain with a level of 
granularity not super high not fine grained, though. The average CPU usage of all 
machines is about 70% and the system is well balanced given that the machines are 
taking more or less the same computational time. I'm using asynchronous communications 
to minimize the communication cost, but even though the communications are taking more 
time than each machine spends in its job. With a coarse grain domain the application 
is working as expected. Since i'm testing the application on a LAN with all the 
machinery with dynamic Ips, i'm using local names. Therefore, my question is: what are 
the expected differences in terms of throughput if one uses remoting with local names 
or ip addresses?

Regards,

Manuel 

You can read messages from the Advanced DOTNET archive, unsubscribe from Advanced 
DOTNET, or
subscribe to other DevelopMentor lists at http://discuss.develop.com.

Reply via email to