Tivoli Enterprise Monitoring Servers - hub and remote

• Retrieves data from the monitoring agents and delivers data to the portal server.
• Sends alerts to the portal server when conditions specified in situations are met.
• Receives commands from the portal client and passes them to the appropriate monitoring agents.

Should you install a monitoring server on a z/OS, Windows, UNIX, or Linux system?

Many organizations prefer the reliability and availability characteristics of the z/OS platform for the monitoring server.

On the other hand, if your installation runs monitoring agents for other platforms, you might prefer a distributed platform such as Windows or Linux for your hub monitoring server. If you install the hub monitoring server on Windows or Linux, you have the option of deploying the portal server on the same system, which can shorten the communications path.

• The hub monitoring server is the focal point for managing your environment. You can configure only one hub monitoring server. It communicates with the portal server, with monitoring agents, and optionally with monitoring servers running remotely.
• You can optionally configure a remote monitoring server to distribute the workload of the hub monitoring server, but it is not required.

Each remote monitoring server must be installed on its own system or workstation. A remote monitoring server communicates with the hub monitoring server and with monitoring agents running on the same or different systems.

Note that a remote monitoring server is remote only with respect to the hub monitoring server, not necessarily with respect to the monitoring agents. A monitoring agent can be installed on the same system as a remote monitoring server. The monitoring server is then local to the monitoring agent, but it is still a remote monitoring server. See “Tivoli Tape Optimizer monitoring agent” on page 10.

The configuration scenarios in this guide assume that the monitoring server being configured with the Tivoli Tape Optimizer Monitoring Agent is a hub monitoring server. For instructions on configuring remote monitoring servers, see the IBM Tivoli Monitoring: Configuring IBM Tivoli Enterprise Monitoring Server on z/OS and IBM Tivoli Monitoring: Installation and Setup Guide.

Should you configure a remote monitoring server or servers for your environment?

A remote monitoring server is designed to offload work from the hub. Whether or not your hub gets overloaded enough to slow down hub processing of situations and other data depends on the complexity of your environment. The following factors tend to boost strain on the hub and increase the likelihood that you might want a remote server to help out the hub:

• Monitoring many z/OS images. The more monitoring agents you have installed on z/OS systems, the more work for the hub.
• Monitoring many situations. Tivoli Tape Optimizer Monitoring Agent does not come with a great many situations to consume hub cycles, so unless you have other monitoring agents with lots of situations, this is probably not the deciding factor.

Configuring a remote monitoring server can also give you scalability potential and failover protection, which might be especially important when you add Tivoli Tape Optimizer Monitoring Agent to an environment with many monitoring agents. For more information on these issues, see the IBM Redbooks: Deployment Guide Series: IBM Tivoli Monitoring 6.2 at the following Web site:

Look for the following topics:

• Small/medium installation
• Scalability