Healthy VoIP Nets�Part VIII�Stocking the Toolbox for the Lower Layers
We herein detail four categories of tools the savvy net manager will have on hand at all times for troubleshooting the Physical layer.
In our last two tutorials, we reviewed the Open Systems Interconnection (OSI) Reference Model, and used it as a framework for comparing the various functions of a VoIP networking environment. In those discussions, we found that the lower three layers are typically implemented in hardware, and the upper four layers are typically implemented in software.
These distinctions suggest that different tools will be required to address implementation or management issues at the various layers, and that the network manager will need more than one tool in his or her toolbox. In this tutorial we will explore the different categories of tools that would be good to have in your network management arsenal, beginning with those that are used to test the Physical layer.
Electronic Hand Tools: At the risk of preaching to the choir, you would not believe the number of clients I have seen over my two decades in the consulting business that did not have some basic hand tools around the office. Some of those that claimed to have tools could not put their hands on them when needed, as "someone must have borrowed and not returned them."
To avoid these frustrations, spend a few hundred dollars on a good electronic toolkit, and keep it locked up where you can find it. A good source for these toolkits that has been around for years is Jensen Tools, now a business unit of The Stanley Works.
Environmental Monitoring Systems: These systems alert the network manager to changes in the network physical environment, such as excessive heat or humidity, smoke or fire, commercial power problems, or potential water damage. In many cases, these issues are the responsibility of the plant department, not network management, but if the air conditioner fails in the server room, expect to get a lot of telephone calls no matter who has the ultimate responsibility. Some of these systems come equipped with wireless sensors that communicate with a base system, and can also alert you via phone, fax, or email for event and status updates.
Infrastructure Testing: Just a few years ago, cable scanners to handle unshielded twisted pair cable was the norm, but now with infrared and fiber optic links between buildings, and wireless networks inside, management tools to cover this expanding network infrastructure may need to be updated. So compare the capabilities of your cable test equipment with the various LAN and WAN links that are currently installed, and make sure that you can cover all of the Physical and Data Link technologies that are currently being deployed.
Network and Performance Monitors: If you live in a major metropolitan area, you likely have at least one local radio or television station with a helicopter that flies overhead during rush hour to alert you to accidents, congestion, or other problems on the freeways that would impede your commute into the office. Those news reports likely tell you the type of problem (e.g., traffic congestion), but not necessarily the details of the problem (e.g., a vintage roadster with a blown lower radiator hose that has overheated in the left lane). And you are likely most interested in avoiding the problem or doing something that might make the problem even worse (like inadvertently taking the same route), rather than hearing all of the mind-numbing details.
In the networking arena, the notification of a WAN link that is down, or a server with an unusually high CPU utilization can be key information. Granted, that simple notification may not solve the problem in the long term, but it provides a short term alert that may enable you to re-route the traffic or system to another network and puts you on the path to a longer term solution. And since we are dealing with real time systems, such as voice, and video, that monitoring process also buys you some much needed time to more fully analyze and correct the problem (which may require net management tools with more sophisticated capabilities.
Network monitors may be either centralized or distributed systems. In most cases, they are relying upon either historical data or user-specified inputs to set their operational benchmarks and alert thresholds. In addition, the monitor's output is in the form of color-coded graphs and charts, which can show at a glance if an abnormal condition exists, and thus quite useful for upper management's perusal.
More sophisticated systems may overlay this information on a map of the network. Monitored parameters may include traffic statistics, call setups/disconnects, failed network access attempts (security breaches), host utilization, server response times, and Quality of Service (QoS) parameters, such as Mean Opinion Score (MOS) results or R-values.
Our next tutorial will continue our discussion of VoIP network management tools.
Copyright Acknowledgement: © 2008 DigiNet Corporation ®, All Rights Reserved
Mark A. Miller, P.E., is President of DigiNet Corporation®, a Denver-based consulting engineering firm. He is the author of many books on networking technologies, including Voice over IP Technologies, and Internet Technologies Handbook, both published by John Wiley & Sons.