Customer X are in the oil and gas industry and operate globally from Aberdeen. They were due to relocate locally to a new head office and at the same time move from a traditional ISDN30 based telephone system over to a more cost effective and resilient IP telephony based solution.
Figure 1 shows the original network, the rig traffic (1000 DDIs) was taken via satellite to a teleport in Aberdeen. It was connected to the customer’s phone system in their head office, this was serviced by 90 channels of ISDN30. There was also the office voice traffic from the head office and a few other remote offices, these were served by their own individual ISDN30 connections totalling around 90 channels and a range of 1000 DDIs.
The switchover to SIP trunks was phased to keep downtime to a minimum. Our first step was to install a single IPDC endpoint with 5 channels, test DDIs and run over a dedicated Ethernet circuit at their current head office. The customer could then configure and test their phone system on the IPDC endpoint using the test numbers, this is shown in Figure 2.
Once the customer had fully tested the IPDC product with their phone system and were happy with how it functioned we then increased the number of channels on the endpoint from 5 to 90. The customer’s rig voice traffic was then migrated from the ISDN30 phone system over to the IPDC endpoint as shown in Figure 3. This included porting all of the DDIs, this was done out of hours to reduce the impact to the customer.
We installed two new 100Mb Ethernet circuits at the customer’s new Head Office, for resilience purposes these were provided from two different suppliers. 20Mb had been reserved on each circuit purely for voice traffic.
We then set up two resilient IPDC endpoints at the head office, one for the rig voice traffic and another for the office voice traffic. The endpoint for the rig traffic was set up in active / standby mode with 90 channels running on each Ethernet circuit. The channels on the secondary circuit are in standby mode, if the connection to the primary circuit goes down the traffic is immediately switched to the secondary circuit.
The endpoint for the office traffic was also set up in active / standby but using the Ethernet circuits in the opposite way i.e. the primary office endpoint is set up on the Ethernet circuit used for the secondary rig endpoint. This means that if one of the Ethernet circuits went down, the other circuit would be taking all 180 channels of voice traffic.
When the customer moved into their new premises, we migrated all 1000 DDIs from the rig traffic IPDC endpoint at their old Head Office on to the resilient IPDC endpoint in their new head office. For the office IPDC endpoint they wanted to use new DDIs so we allocated a range of 1000 numbers and their old head office ISDN30 was ceased. There was also data in use on the remaining 80Mb of each of the circuits but the purpose of this document was to highlight how the voice changeover was managed. Their current set up is shown below in Figure 4.