VoIP performance evaluation and monitoring methods

The debate over VoIP is beginning to settle, and best practices for proper VoIP operation are beginning to take shape. A well-defined camp has been formed. Some have methods for evaluating and monitoring networks, and some have interesting hybrid technologies. You don't have to worry about anything that is quite different. Now you can find out which technologies are suitable for your needs.

Network management companies have always viewed the VoIP market as a place to expand and generate more sources of revenue. Some companies have repackaged existing technologies and added the functions and capabilities that are very important for voice communication. Some companies only slightly modified the user interface and simply changed the name. In many cases, unique technologies have been developed to replace those reliable methods.

Whether you are looking for some old, new, borrowed, or IBM technologies, the real question is whether it is suitable for your application management needs. Sometimes this is hard to say. In order to attract the attention of the market, there are a lot of practical propaganda and continuous nonsense. Everyone adapts to this situation.

Therefore, how do you differentiate various products without worrying about whether these products can meet your needs?

Regardless of how the vendor promotes how it is different, there is a feasible way to divide most solutions into two main categories. These classes may include some interesting variants, but most methods are suitable for them (with some exceptions, I will discuss it later ).

Two main technical methods of VoIP performance evaluation and monitoring include:

1. Traffic Simulation-the synthesis of VoIP or VoIP-like communication is actively transmitted between two nodes or proxies. Analysis of these communication produces a description of the network path.

2. Traffic Monitoring: passively monitors the actual VoIP communication at one point in the network for real-time and historical analysis. These analyses describe the performance through that point.

Of course, each method has advantages and disadvantages. Here you will see where these methods are suitable for your current network management process.

Traffic Simulation

Place software agents or hardware devices on two or more points in the network and run the combined VoIP communication between these points. The tested data can be intensive data in a short time.

(For example, used for pre-application evaluation), you can also monitor performance without interference for a long time. The generated data can simulate a large number of telephone loads, encoding types, and VoIP. This kind of traffic simulation is also

It also includes communication with VoIP devices such as mobile phones or exchange servers.

Advantages:

◆ Strictly controlled test methods

◆ Able to execute and monitor important network paths (such as WAN connections) in detail)

◆ Be able to test extreme conditions (stress testing) in advance)

◆ Provide very accurate MOS prediction, with the best/average/worst range

◆ Pre-application evaluation supported

Disadvantages:

◆ A proxy or hardware device is generally required for each node.

◆ Unsatisfactory application overhead and maintenance problems often occur

◆ This method is not available if you cannot access the remote network.

◆ This method will put a lot of pressure on the network

◆ Generally, the troubleshooting work for "on-demand application" is not effective.

◆ MOS with poor performance can be found, but the cause or location of the problem cannot be found.

◆ The end-to-end path is not included. The last mile or specific device performance may be ignored.

1. How to Implement VOIP detection and real-time monitoring
2. Call Control for enterprise VoIP networks