Got Dial Tone?

In the early days of VoIP adoption, enterprises learned a great deal about the planning, implementation, and ongoing network infrastructure management required to ensure the reliability and QoS of basic voice service delivery. Dial tone reliability is fundamental to business continuity. Many lessons continue to be learned as VoIP users discover that modifications to routers and switches, updates to server configurations, and changes to any part of the distributed networking infrastructure can now impact the availability and performance of voice communications.

When you add custom voice application development to the infrastructure issues, reliability concerns become even more serious. Early adopters have encountered a number of problems. For example, many IP telephony protocol implementations are not yet mature, and while many vendors utilize standards such as H.323, MGCP, or SIP, they often force those standards to work within their environments by adding proprietary extensions. Additionally, early developers have found that many APIs and protocols are very unforgiving, and errors in their use can result in crashed IP phones, or worse, adversely affect call processing on the switch. Even to the extent that the protocols work correctly, improper or inexperienced use of the protocols can easily impact voice availability and performance.

The core problem underlying these kinds of issues is that IP PBXs often expose development interfaces, but they simply do not provide the robust development capabilities expected of an application platform. Custom applications are running directly against the PBX. There is no middleware to buffer and protect the IP PBX from any variety of ad hoc developer access and use. There is no way to ensure that the applications will not impact basic dial tone and core call processing.

Learn how Metreos addresses these issues with the Metreos Application Runtime and Metreos Media Engine.