One think you to consider. NCM does not poll at the same rate as NPM. In particular NCM is only pulling data when either a job kicks off to pull configs or if you have real time config change notification turned on then it pulls a config right after the trap. So from a processor load there is very little continuing effect on the process. In fact we have a main poller and 5 additional pollers and prior to the upgrade we only had NCM running on the main poller. So for us the new versions spreads that out across all the pollers
↧