FRAMINGHAM (07/03/2000) - Network probes are small hardware devices that are designed to passively collect data. This may seem mundane, but the data these probes collect provides information that is vital to your enterprise network.
Probes provide the mechanism needed to gather data essential to the construction of network history and trends. This information lets network professionals make timely, informed decisions during a network crisis or failure, and even take action to prevent a problem. This data also provides a manager with key information when it becomes necessary to upgrade or enhance a network.
Today's networks are complicated entities that must be managed properly. You can't manage what you can't see, however, and network probes provide that visibility.
Other approaches, such as SNMP or embedded remote monitoring, do not provide enough accurate, detailed information because they are designed to view one device's interface or port at a time and don't provide an integrated view of the whole network.
A good probe provides accurate statistics concerning a network's operation, such as traffic analysis and top users, applications and protocol usage. Using this information, you can manage the network efficiently, ensuring peak operation and performance.
It is important to utilize the correct type of probe. Many probes collect only simple measurements or don't have capacity to store enough collected data for meaningful analysis. You should select a probe that is capable of collecting data at a level of detail that will indicate the true overall performance of the link or segment of the network being monitored and that has enough memory to store data for accurate historical analysis.
Also make sure a probe is capable of collecting data even at the network's near-saturation point, when visibility is most crucial. Some probes collect data only up to 30 percent utilization of the network, leaving you vulnerable when your network is in peak use.
Some probes claim to work with several types of network media. These probes generally do not provide enough data specific to the media being monitored.
A probe that claims to measure either V-series circuits, frame-relay and/or ATM links will most likely provide less data than a probe designed for just one specific media. If you are measuring ATM links, choose a probe that was designed specifically for ATM that tracks and reports on areas such as permanent virtual circuits, connection usage, switched virtual circuits, and physical layer and ATM Adaptation Layer-5 usage and errors.
Placement is the most important factor in gaining maximum value from probes.
For example, the location of an ATM or WAN probe can provide quality-of-service information that would otherwise be lost.
With these links, it is imperative you place the probe at the first physical point where the circuit enters your network. This ensures that you capture actual traffic on the link before it is terminated in the network DSU/CSU. The DSU/CSU consolidates transmission data, eliminating the granularity needed to make accurate assessments of the network's performance. If you place the probe after the DSU/CSU, you will never see the errors on a link, reducing its overall efficiency.
A probe before the DSU/CSU can, for example, provide you with information to ensure your network vendor is meeting the level of quality promised. Having a probe placed before the DSU/CSU in a frame-relay network gives insight into whether your network vendor was providing you the correct committed information rate in an error-free manner.
In addition, probes can work hand-in-hand with network management software to collect vital statistics concerning a network's overall performance.
IT managers will need greater visibility into their enterprise networks to pinpoint potential problems before they effect performance and hamper e-business. Probes provide that necessary visibility.
Cooper is a regional engineer at Agilent Technologies Inc. NetMetrix Division in Palo Alto. He can be reached at firstname.lastname@example.org.