Monitoring Across an Enterprise

I l @ ve RuBoard

There are problems with trying to manage a large number of systems. Typically, an IT department will be responsible for managing multiple systems, most likely on multiple platforms. Big companies have a large number of systems that need to be managed.

Keeping track of multiple systems on multiple platforms can be a complex task. Some companies want to implement a lights-out environment, in which operators are not required at every computing location in the enterprise. Other companies require the availability of systems 24 hours a day, 7 days a week. Most companies also have a disaster recovery plan to protect their critical data and application resources. These are some of the issues you must deal with when managing an enterprise of computer systems and applications.

Several solutions are available to aid in enterprise management. For example, HP Open- View IT/Operations (IT/O) is one product designed to help with enterprise management. It provides an Event Browser and a topology map with different icons to represent all of the various systems in an environment. IT/O uses different icons on the map to represent different platforms. For example, there are distinctive icons used to identify NT systems and HP-UX systems.

Using a standard like SNMP from IT/O, you can receive events and poll heterogeneous systems in the enterprise using a common interface. IT/O allows you to distribute monitoring and message templates to multiple systems. It is scalable, so you can easily add additional systems to be managed. To aid in event management, you can integrate one of several event correlation tools to coordinate and reduce the number of messages coming in from systems in the enterprise. IT/O also supports a sophisticated architecture in which events can be forwarded, with annotations, to another IT/O management station.

For disaster recovery, many companies have sophisticated procedures in place and practice disaster recovery annually. One thing to consider is a backup plan for the enterprise management platform itself. Many of the enterprise management frameworks discussed in this book have been certified to run in an MC/ServiceGuard environment, meaning the whole enterprise management framework can start up on an alternate node in the event of a system failure.

When managing multiple systems in an enterprise, using common tools and standards can greatly reduce the number of tools you need to learn. Using a centralized console reduces the number of places you need to look to check for problems. Enterprise monitoring frameworks are important for reducing the complexity of monitoring systems in the enterprise.

I l @ ve RuBoard


UNIX Fault Management. A Guide for System Administrators
UNIX Fault Management: A Guide for System Administrators
ISBN: 013026525X
EAN: 2147483647
Year: 1999
Pages: 90

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net