 |
New Study Indicates Strong Concern Over VoIP Reliability
During the past year we questioned nearly 300 network engineers about the challenges they faced in managing network performance and VoIP.
In an analysis of surveys from network engineers across the United States, the study found that:
Nearly 50 percent were concerned with their ability to monitor the quality of VoIP service.
41 percent were unsure of their network’s ability to handle the extra bandwidth consumption from VoIP calls
36 percent were concerned with the reliability of their VoIP application during periods of heavy use.
Read the study |
|
In Practice: Avoiding Problems Through VoIP Pre-deployment Testing
Summary
VoIP is a complex technology to deploy with many potential bumps along the way. Like the adage “measure twice, cut once,” it’s better to avoid network headaches by spotting potential issues during pre-deployment testing rather than assuming your network can handle the new VoIP application. |
In our study, nearly half of the organizations indicated they use VoIP, and another 30 percent plan to implement the technology in 2007. Given the high percentage of VoIP implementations, the amount of concern network engineers expressed about technology performance was very surprising. Nearly 50 percent of network engineers were concerned with their ability to monitor the quality of VoIP service. 41 percent were unsure of their network’s ability to handle the extra bandwidth consumption from VoIP calls.
Most of the challenges expressed by companies in implementing VoIP were caused by a lack of proper equipment or planning. Often these organizations implemented VoIP without realizing how susceptible the technology was to network conditions. They also were surprised by the high expectations users had for VoIP performance. |
 |
Fortunately, there are several steps network engineers can take to ensure VoIP technology meets everyone's expectations. Pre-deployment testing is the first step toward ensuring the network is ready to bear VoIP traffic. Monitoring and measuring VoIP performance is vital to removing any unforeseen obstacles before actual deployment and ensuring a successful rollout.
Testing before deployment should include not only the VoIP application and the infrastructure, but it should also review the performance of other applications running concurrently with VoIP. After the testing process, IT staff should be able to:
- Determine the amount of VoIP traffic the network can handle without audio degradation
- Discover any configuration errors with the network and VoIP equipment
- Identify and resolve sporadic problems that affect network and application performance
- Identify security holes that allow malicious eavesdropping or denial of service
- Guarantee call quality matches user expectations
Relying on the default testing tools included with the VoIP equipment is often not sufficient for successfully testing and managing VoIP. These tools provide minimal insight into VoIP communication running over the network. Beyond limited testing capabilities, default testing tools typically fail to compare VoIP performance with overall network performance, offering a very limited view for troubleshooting and management purposes.
Enterprise-level VoIP management requires tools from an independent third party to test and monitor VoIP. Having a protocol analyzer with VoIP monitoring and analysis capabilities, such as Observer® Expert, provides detailed insight into both VoIP and overall network performance. Observer can identify and resolve audio problems, offer real-time call detail records, and display VoIP “Connection Dynamics."
The insights on network and application performance gained from pre-deployment testing and continuous monitoring of the added VoIP traffic will help network engineers to establish benchmarks for acceptable network and VoIP application performance. These benchmarks help the engineer to intelligently configure alarms on the monitoring tool to notify them when VoIP performance deviates from the norm—allowing the administrator to become aware of problems as they arise and resolve them immediately.
When implementing VoIP on a network, there are many potential pitfalls. The good news is through pre-deployment testing many of the pitfalls can be easily avoided. The information gained from testing will also help establish benchmarks to guarantee the long-term success of VoIP in your organization.
Tech Tip: Setting Server Configuration
It’s important to have your VoIP servers properly configured in Observer Expert under the Expert VoIP Settings menu to correctly identify VoIP traffic and connections. If the server IP and type are not properly configured, it may be difficult for Observer to identify when calls were established (call setup) and completed (call tear down).
For example, Avaya VoIP systems use specialized media processors and control processors to setup and tear down calls. Without these settings configured, a call transferred from one phone to another phone may not be followed, leaving Observer to assume that the original call timed out. It will then establish the transferred call as a new call. Observer offers an easy way to identify these servers to help with call identification.
- Choose Capture > Packet Capture from the Observer main menu.
- From the Packet Capture screen click the Decode button at the top.
- From the Decode and Analysis screen select the Expert Analysis tab in the lower left-hand corner. Then select VoIP Events on the left side listed under Expert Data.
- Click on Settings within the VoIP Events menu. This will cause the Expert VoIP Settings menu to appear. Under the General tab, you can list specific servers in the white box below Server Configuration.
- In the white box, click below IP to list the address of the server. You can then choose the IP type from the drop-down menu indicating it as a Server IP, Outside IP, Media Processor, or Control Processor.
|
 |
March 2007 
