The SkySwitch network consists of multiple redundant server clusters hosted in New York, Dallas and Las Vegas. All of our servers operate in an Active-Active configuration, which means that the service is available from either cluster at any time. A further level of redundancy is achieved by using a "hot spare" for each critical server. In the event of a hardware or software error, the spare server will assume the IP address of the server with a problem, and assume its operations transparently.
The SkySwitch device provisioning servers allow you to Provision any number of SIP devices remotely, without having to manually configure handsets. The device provisioning servers are available at http://sipcfg.io. We use geographically-aware DNS to point users to the provisioning server that is best able to respond to a request.
The primary web User Interface for SkySwitch, called the Manager Portal, is available at http://pbx.skyswitch.com. We use geographically-aware DNS to point users to the web UI that is best able to respond to a request.
All communication between Registered Devices and/or SIP Trunks and the SkySwitch network will be through the a primary IP address in NYC and DAL. Upon registration, the SkySwitch provisioning server will configure SIP Devices to register to the SkySwitch SIP server with the lowest network latency (if you would like to override this assignment, please contact support with your request). These servers will be the only ingress and egress points for registered SIP devices.
Behind the scenes there are many other SIP Servers at work. These include Session Border Controllers, ENUM Gateways, Ingress Proxies, Egress Proxies and Conferencing Servers. These are largely hidden from your view. However, you will be better able to understand SIP Call traces if you are aware of the other elements that fit into a given call flow. The diagrams below show the typical call paths taken for Inbound and Outbound calls from a registered SIP device.