OpManager Doesn't Scale Well
We have been using OpManager for a few months now, and the CPU utilization on our server is through the roof. This product really needs to start using host based agents on the remote machines to send data back to OpManager. All of the larger product suites (OpenView, Altiris, etc.) do this for the simple reason that the monitoring host will get overloaded the large the implementation becomes.
We are monitoring about 140 servers and about 20 routers and switches so the polling is quite intense. OpManager is poling over 64,000 items during these polling cycles. This model is simply way too inefficient for large scale data center monitoring. There are also a lot of inconsistencies between product lines. We have seen a host based agent in Security Manager plus, but where is it in the product that desperately needs it?
Please advise if we will be seeing an agent we can install on monitored systems in order to offload CPU cycles to the host.
Regards,
E-
We are monitoring about 140 servers and about 20 routers and switches so the polling is quite intense. OpManager is poling over 64,000 items during these polling cycles. This model is simply way too inefficient for large scale data center monitoring. There are also a lot of inconsistencies between product lines. We have seen a host based agent in Security Manager plus, but where is it in the product that desperately needs it?
Please advise if we will be seeing an agent we can install on monitored systems in order to offload CPU cycles to the host.
Regards,
E-