The discrepancy between network monitoring and management has blurred now that technology, service and labor changes have taken place in the enterprise. This change requires an update to your WAN network monitoring strategy.
How a WAN network monitoring strategy has changed
Network connections are the fabric that delivers IT services and supports collaborative relationships in the enterprise, and any failure in the network is likely to cut users off from critical applications and destroy team productivity. From the onset of enterprise networking, a critical requirement for network operations managers and planners has been to monitor network behavior to detect trends that could compromise performance and to respond to problems proactively. This requirement remains real today, but the way it is addressed is evolving under the pressure of technology, service and even labor changes.
One reflection of the evolution of monitoring is its collision with network management. A WAN network monitoring strategy has always required the analysis of traffic in real time, but most management systems support at least some form of real-time traffic analysis too; so the distinction between monitoring and management may become more cloudy. As network management systems improve, in fact, monitoring has tended to focus on areas where general statistical information about network performance may not be enough.
What should your WAN network monitoring strategy accomplish?
Most enterprises link WAN network monitoring with one or more of three specific missions:
- Application monitoring: Monitoring at the application level to measure the service quality actually delivered to the application. This is growing out of a need by network and IT management to provide specific performance guarantees to line departments they serve.
- Network device monitoring: Monitoring at the device level to determine the health of network components and the extent to which their performance matches capacity plans and intra-enterprise service-level agreements (SLAs). This is the primary focus of network monitoring for enterprises with large network investments.
- Boundary monitoring: Monitoring at the trunk/connection level to detect faults, manage SLA commitments by WAN service providers/vendors, and troubleshoot in the case of failures. This "boundary monitoring" is on the rise as enterprises adopt increasingly complex IP and Ethernet services whose SLA terms are more complicated to verify.
What form should your WAN network monitoring tool take?
For each of these WAN monitoring strategies, there are two high-level options:
- A device or interface's management information base (MIB) can be monitored.
- A probe can be used to gather specific information at a lower level, perhaps even independent of the devices involved.
Ten years ago, network management tended to be based on MIBs and on the Simple Network Management Protocol (SNMP), while network monitoring tended to be based on proprietary probes and a separate set of central tools to analyze the data.
The desire to centralize all network management/monitoring tools into a single cohesive application set has converged monitoring and management at the network operations level. As enterprises have moved to deploy management strategies, they've adopted a strategy of using a high-level console to display network health overall, with the ability to drill down to isolate specific problems. HP's OpenView platform is an example of a management console that can integrate both high-level management and lower-level monitoring tools.
Continue reading this tip in part 2: WAN network monitoring and management are one.
About the author:
Tom Nolle is president of CIMI Corporation, a strategic consulting firm specializing in telecommunications and data communications since 1982. He is the publisher of Netwatcher, a journal addressing advanced telecommunications strategy issues. Check out his SearchTelecom.com networking blog Uncommon Wisdom.