The first step in troubleshooting network congestion is to note the times when the congestion occurs. Try and measure...
network traffic over time, so that you have a baseline of activity during low and high usage periods. The highest usage period is typically the beginning of the day, and the beginning of the week. Other high usage times are just before a company's workday ends. You should also ascertain whether the congestion is a recent issue, one that has been building up over time (say because your company added more employees), or is a singular spike event.
You can use a protocol analyzer to determine which segments in your network have the highest level of traffic, in order to locate potential bottlenecks. Congestion at the edge of the network often indicates that the network has a poor design, something that may require the upgrade or creation of a more robust network backbone.
One issue to check with is whether there is a large number of packets coming from one or a small number of IP addresses. The common cause of this problem is a malfunctioning network card. Replacing that card can solve this kind of problem. If you find that an application is responsible for generating a lot of packets you may want to check if there is a service in that application that is responsible and can either be tuned or turned off. You may also want to split the application activity among more or differently located servers.
The number of protocols you use on a network can also be an issue. Whenever possible, reduce the number of protocols in use, and avoid broadcast protocols such as NetBEUI. WINS is one way to reduce NetBEUI, but even its usage has given way to NetBIOS over TCP/IP, which eliminates the need for NetBEUI altogether.
Barrie Sosinsky is president of consulting company Sosinsky and Associates (Medfield MA). He has written extensively on a variety of computer topics. His company specializes in custom software (database and Web related), training and technical documentation.