8.6 Summary

Our study of network management, and indeed of all of networking, is now complete!

In this final chapter on network management, we began by motivating the need for providing appropriate tools for the network administrator - the person whose job it is to keep the network "up and running - for monitoring, testing, polling, configuring, analyzing, evaluating and controlling the operation of the network.  Our analogies with the management of complex systems such as power plants, airplanes, and human organization helped motivate this need.  We saw that the architecture of  network management systems revolve around five key components - (i) a network manager, (ii) a set of managed remote (from the network manager) devices, (iii) the management information bases (MIBs) at these devices, containing data about the device's status and operation, and (iv) remote agents that report MIB  information and  take action under the control of the network manager, and (v) a protocol for communicating between the network manager and the remote devices.

We then delved into the details of the Internet Network Management Framework, and the SNMP protocol in particular.  We saw how SNMP instantiates the five key components of a network management architecture, and spent considerable time examining MIB objects, the SMI - the data definition language for specifying MIB's, and the SNMP protocol itself.  Noting that the SMI and ASN.1 are inextricably tied together, and that ASN.1 plays a key role in the presentation layer in the ISO/OSI seven layer reference model, we then briefly examined ASN.1.  Perhaps more important than the details of ASN.1 itself, was the noted need to provide for translation between machine-specific data formats in a network.  While the ISO/OSI reference model explicitly acknowledges the important of this service by the existence of the presentation layer, we noted that this layer is absent in the Internet protocol stack.  Finally, we concluded this chapter with a discussion of firewalls - a topic that falls within the realms of both security and network management.  We saw how packet filtering and application-level gateways can be used to provide the network with some level of protection against unwanted intruders, perhaps allow the network manager to sleep better at night, knowing the network is relatively safe from these intruders.

It is also worth noting that there are many topics in network management that we chose not to cover -  topics such as fault identification and management, proactive anomaly detection , alarm correlation, and the larger issues of service management (e.g., as opposed to network management).  While important, these topics would form a text in their own right and we refer the reader to the references noted in section 8.1.