VoIP

Problem Overview

VoIP issues are highly visible and can have serious impact on the day to day running of a business. VoIP infrastructure in large organizations is complex and distributed, and voice traffic is extremely latency sensitive. Compounding the problem, VoIP traverses the network as one of many competing applications and voice signalling and call quality can be adversely affected by interaction with other network traffic.

According to Gartner: “Real-time voice and video are the most network-sensitive applications. When the network performance is challenged by congestion, these typically break first.” In the same report, Gartner indicated that a wire data approach to VoIP management was “desirable” in medium sized organizations, and “essential” in large organizations.

Solution Overview

Corvil provides real-time reporting and troubleshooting of performance issues for VoIP services. This includes tracking the full lifecycle of every call from setup to teardown and reporting on real user experience based on MOS scores derived from actual RTP traffic seen on the network. MOS scores are reported based on jitter and loss measurements of 100% of RTP media traffic for up to 15,000 concurrent calls in a single appliance.

Corvil passively taps into the packet data flowing in the network - this is the only way to get an accurate representation of VoIP activity. The Corvil solution uniquely mines this packet data by automatically discovering all VoIP calls, tracking the lifecycle of the call and reporting the key performance metrics in real time. The user can search on a handset ID and see all calls, or start with a call and see all signalling messages involved in the call. The packet record is only a single click away, and you can quickly export the packets for call replay.

Outcomes

• Live production dashboards of all calls in progress, MOS scores, gateways volumes and performance, signalling paths
• Reduce mean-time to repair of VoIP escalations – one click access to the underlying packets
• Scalable to report real user experience for 15,000 concurrent calls per appliance
• Cost effective VoIP packet capture on remote network segments for VOIP replay
• Hold Service Providers to account by reporting point-to-point latency, jitter, loss of every VoIP call

Voice Quality Metrics - RTP MOS, Jitter and loss based on measuring 100% of network packets for every VoIP call.

Voip Signaling Errors - Report on call setup times and signaling errors for the full breadth of VoIP protocols including some vendor variants.

Improper Qos Marking - Automatically discover and report on all RTP media packets that have not been marked as EF on the network.

Concurrent Calls - Discover and report the maximum concurrent call active at any signaling gateway/SBC.

Quality Analysis By Codec - Report MOS scores per discovered Codec active in the network.

Service Provider Performance - Measure the point to point latency, loss and jitter for every RTP media packet as they traverse your service provider WAN links.

QoS Analysis - Analyze queuing and configure appropriate bandwidth for the VoIP priority queues.

Signaling Tcp Analysis - Report on VoIP signaling session errors caused by network retransmissions or TCP zero window events.