HP Operations Manager for Windows

Deployment models


There is no fixed upper limit for the number of nodes you can monitor within an HPOM environment. Scalability of agent-based or agentless monitoring in your environment primarily depends on the resources on the management server system, the managed nodes, and the network.

The effect agentless monitoring has on the CPU and memory utilization on the proxy system is dependent on the number of nodes, the number of metrics that are gathered and the frequency of gathering that is used. The network connection and bandwidth between the proxy and the monitored nodes can also influence the time needed for gathering the data.

Therefore, we recommend that you monitor resource utilization on the proxy system to avoid problems with other applications running on the proxy. If monitoring significantly affects other applications, reduce the gathering frequency, choose another proxy system with more resources or split the work among several proxy systems, sharing the monitoring of nodes and metrics, as described in Proxy systems in large environments

The location and arrangement of proxy systems in agentless monitoring can be divided into the following scenarios:

HPOM management server system as the proxy system

The most simple arrangement is to use the HP Operations agent running on the HPOM management server system as a proxy agent for remote monitoring, that is the HPOM management server system also acts as a proxy system.

Note NOTE:
If you plan to install additional HP BTO Software products, such as HP NNM, HP Service Desk, HP Performance Manager, HP Reporter, or HP Internet Services on the system hosting the HPOM management server and proxy agent, make sure that the available system resources are sufficient for all installed products. Resource consuming processes reduce the performance of remote monitoring as well as the performance of the HPOM management server and other installed software.

The following recommendations apply:

Independent proxy system

Using an independent proxy system is recommended when the HPOM management server system does not have sufficient resources to handle the proxy tasks without a noticeable impact on the performance of the HPOM management server. In this case, select a suitable managed node to play the role of the proxy system.

The following recommendations apply:

Proxy systems in large environments

Whenever the number of agentless nodes being monitored is a strain on a single proxy system, we recommend using additional proxy systems.

For each proxy system, the following recommendations apply:

Note NOTE:
Any combination of the possible proxy system configurations is acceptable, that is, you could have one proxy system being the HPOM management server system, and one or more proxy systems being any other suitable managed node.

Related Topics: