Chapter 9: Network Management

In the old days of small work group LANs, it was relatively easy for a system administrator to keep tabs on the status of desktop PCs, servers, and the network simply by looking at the lights on the front of the equipment. As these networks grew in complexity and scope, it became more than any person, or group of people, could do to know the status of all parts of the network at all times. This problem provided the challenge for the first network management system (NMS). The early NMS software was little more than a log reader, similar to the Event Viewer in Windows Server 2003 today. Next, the ability to read status and alert messages in a standard format was added. This standard format became the Simple Network Management Protocol (SNMP). Manufacturers quickly added the ability to format and send SNMP messages to all of their equipment. Today, virtually all network infrastructure devices such as routers, switches, bridges, and CSU/DSUs, as well as servers and operating systems, can report their status using SNMP. It is this capability that makes modern NMS packages like Microsoft Systems Management Server (SMS) and Microsoft Operations Manager (MOM), Citrix Resource Manager (RM), and HP OpenView possible. The ability to receive and collate SNMP messages is only the tip of the iceberg of what an NMS can do and what your organization should use it for.

Although server-based computing is, by nature, more centralized and architecturally simpler than distributed computing, this does not mitigate the need for a strong system management environment (SME). It is even more critical to establish service level agreements for services delivered and to use tools, such as an NMS, to manage them. This chapter discusses general SME messaging standards, SME characteristics including monitoring and reporting for server-based computing, and concepts for SME implementation using tools from Microsoft, Citrix, Hewlett-Packard, and others.

People, Processes, and Product

Utilizing an NMS is only part of an organization's overall SME. An SME consists of the people, processes, and product ("three Ps") within an organization that effectively manage the computing resources of that organization. "Product" is more accurately "technology," but "two Ps and a T" doesn't have the same punch as "three Ps." We find the simplest way to think of the interrelationship between the three Ps is in terms of service level agreements (SLAs).