VoIP presents the worst possible scenario for network-influenced application performance:

• Highly sensitive to IP behaviors, VoIP will not tolerate even the most rudimentary amounts of latency, loss, congestion or jitter.
• A critical application, the phone is a mandatory tool within an enterprise. Every second someone picks up a phone and just expects it to work.
• With its user-subjective metric, network problems that may go unnoticed now, will cause poor call quality and dropped calls — conditions anyone within the organization can recognizing; including the CEO.

Basically, networks aren’t ready for VoIP. And here’s proof.

• Gartner analyst Jeff Snyder says that 85% of network as they exist today won’t support voice-over IP.
• Based on a survey of 9,900 customer tests, we found a serious network issue 38% of the time:
• 20% of networks have bad NIC card drivers
• 29% of devices have packet loss, caused by:
  • 50% high utilization
  • 20% duplex conflicts
  • 11% rate limiting behaviors
  • 8% media errors
  • 8% firewall issues

While all of these issues pose application performance problems that are poor for data, they spell death for VoIP. So, if practices relied upon for managing data networks in the past fall dramatically short of the needs of the converged network environment, what’s a person to do?

A successful VoIP implementation requires a reliable and accurate network analysis and performance technology and a three-phase approach: network readiness assessment, break/fix troubleshooting and ongoing monitoring and maintenance

Solutions:

• Pre-deployment testing to understand the condition of the network before VoIP is deployed. This ensures that the network is capable of handling VoIP and if not, identifies the faults that need to be repaired to meet performance qualifications. This additional knowledge, in turn, lowers the probability of complaints, churn rates, escalating costs and finger pointing.
• Continuous monitoring and troubleshooting of the network in real time during and after deployment to reduce costly outages and degradation. Network operators can isolate potential problems before they occur.
• Better understanding of network capabilities. With this, service providers can offer guarantees to different service levels (SLAs). This translates into better market response in terms of price, quality, and availability.
• Monitoring of the network as a service. Network managers can view and understand end-to-end performance and rapidly detect, diagnose and control unusual events.
• Designed to meet VoIP’s unique needs, NCT will provide:
• Extensive Coverage: Assesses the entire end-to-end network, including inspecting carrier’s infrastructure and to the organization’s handsets, eliminating risks associated with sparse sampling and guesswork.
• Comprehensive Measurements: Measures bandwidth, utilization, jitter, loss, latency as well as best, actual and worse MOS making it easy to gauge the feasibility of the network to handle VoIP.
• Unmatched Diagnostics: Locates and identifies network configuration faults that impact VoIP call quality. Isolating existing problems and identifying the cause are vital steps to remediation.
• Automated Analysis: Quickly and easily identifies all the hops on the subnet, generates tests, isolates faults and creates reports for later review.

In the end, there is a positive side to VoIP and all this renewed network attention., once the network is conditioned to support VoIP, all applications will achieve higher performance.