Archived
This forum has been archived. Please start a new discussion on GitHub.
question about icegrid
we all know ,there are four load balancing types in ICEGRID(ICE3.2.1),but I am puzzled about the adaptive,it said that Adaptive load balancing uses system load information to choose the least-loaded object adapters over the requested sampling interval.
I want to konw what is the system load information based on ?? is it based on the number of services in node,or somewhat? and ,can i regulate the system load statistics.
another question,can I control the interaction between client and server,for example ,a client request a service,I hope that during the next 30 sencods ,the client can not request the service,and after 30 seconds,the client can request the service again.
I want to konw what is the system load information based on ?? is it based on the number of services in node,or somewhat? and ,can i regulate the system load statistics.
another question,can I control the interaction between client and server,for example ,a client request a service,I hope that during the next 30 sencods ,the client can not request the service,and after 30 seconds,the client can request the service again.
0
Comments
-
Hi,
On Unix, the adaptive load balancing policy is based on the load average of the nodes. On Windows, it's using an average of the CPU utilization. See this section in the Ice manual for more information.
There's currently no way to have a better control over the server endpoints which are returned to a given client (other than the defined load balancing policies) and do something like you suggest. If you have a commercial interest in such a feature, please contact us at info@zeroc.com.
Cheers,
Benoit.0
