Bandwidth from one cable can be risky

Rudolph Muller

Rudolph Muller is the editor at MyBroadband and covers telecoms and broadband news. Rudolph comes from an academic background, but left the University of...

Full Profile

E-mail

Seacom experienced a sub-marine cable failure at 09h19 on 5 July 2010, resulting in service downtime between Mumbai and Mombasa. Early investigations indicated that a repeater had failed on segment 9 of the cable which is offshore to the north of Mombasa.

This unexpected failure affected traffic towards both India and Europe, but traffic within Africa was not affected. Seacom initiated emergency repair procedures to replace the repeater. The repair ship was deployed to the location of the fault to pick up the cable.

Whilst such a repair process should only take a few hours, the overall process may last from 6 to 8 days. This time, the actual duration was unpredictable due to external factors such as transit time of the ship, weather conditions and time to locate the cable. The cable came back into service on 23 July 2010.

To the credit of Seacom, within two days the company had successfully secured a number of restoration options for its clients through other international connectivity providers, including cable networks. This effectively provided their customers with alternatives to reroute services and restore connectivity.

To ensure continuity of service users of sub-marine cable should enter into a restoration contact which provides for the switch-over to another cable should a break in service occur on the main connection route. While Seacom has built-in redundancy at the cable stations on shore and spare fibres in its cable, a repeater failure or a total break will cause total loss of service.

Users of broadband connections on sub-marine cables should enter into a restoration contract to secure at least a fully duplicated amount of capacity to reroute traffic on another cable system. Restoration in most instances is not automatically switched as more than one cable system uses the same facilities.

So although dedicated hard wiring is in place, some additional software and manual patches are required when restoration is implemented. As per the Universal Restoration Manual, it is international practice to attempt restoration on this basis within one hour.

While some Seacom customers had restoration contracts in place with alternative suppliers, many did not and were dependent on Seacom’s ability to secure alternative routing. The deployment of restored capacity via other cable systems progressed very well and Seacom said the company was pleased with the positive support received from clients and suppliers alike.

On 16 July 2010 Seacom reported that during the previous week the designated ship had been deployed to the location of the fault where it proceeded to locate the cable on the seabed. This is easier said than done as the cable would have moved around the seabed from where the installation ship dropped it over a year ago.

This is normal and caused by the natural sea bed movements and strong submarine currents in that area. The faulty section of the cable was at one of the deepest points along its route, some 4700 m below sea level

Once located with the help of cable station staff and specialised equipment, the cable was picked up before being brought onboard the ship to undergo the necessary repair.

This included cutting the cable to remove the affected portion, reviewing the physical integrity of the cable before proceeding with the initial splicing of the spare cable segment which had been tested and prepared while in transit onboard the ship. When the final splicing was completed, the entire system was thoroughly tested before the cable was lowered back in the water.

On 22 July Seacom confirmed that the physical repairs to the fault on the cable were in the final stages of completion, with the entire system undergoing testing before the cable were lowered back into the water.

Technical teams started working with all customers to reinstate their Seacom traffic to pre-outage configurations and all connectivity going out of Africa was fully restored on 23 July 2010.

Although the next day the entire network was fully operational, technical teams continued to work closely with customers to reinstate their Seacom traffic to pre-outage configurations whilst an extensive investigation was taking place to determine the exact cause of the outage.

While sub-marine cables can be risky, restoration is usually reasonably fast. When compared to the possibility of a satellite being taken out by a solar event, it will still be a safer option.

Hats off to Seacom for their fast response in arranging alternative routing and their open communications to keep customers and the media informed of progress in repairing the cable.

Redundancy with undersea cables << comments and views

EngineerIT