Can undersea cable wear and tear punch holes in our image as the world?s backoffice? Fortunately, we weren?t completely at sea when the undersea cable cut off the Egyptian coast recently, thanks to our geographical advantage of two oceans. But the incident has compelled the industry and the government to have a re-look at our disaster management plans, and create foolproof backup in the event of an emergency.
Internet connectivity in India faced disruptions for over 10 days before the cables were repaired and the links were fully restored. The breakdown affected international long distance and internet traffic to Europe and the US, which outsource their back-office operations to India. While big BPO firms saw less damage, smaller firms had a tough time in the absence of foolproof back-up plans.
Suddenly, businesses realised the need to check disaster recovery plans of their telecom service providers in great details. ?There were no connectivity issues experienced by any of BT?s customers in India. I am given to believe ISP?s did report delays shot up to 1,000 msecs,? asserts Sudhir Narang, managing director, BT India. Reliance Communications? subsidiary Flag Telecom, which operates the world?s largest private undersea cable system, spanning 65,000 route kilometers and four continents, says ?Immediately after the cut, Flag Telecom had completed the restoration of services on alternate routes for all customers opting for pre-planned restoration services.? Industry analysts peg the cost of laying a 3,500 km submarine cable at $200 million, while the repair of such faults may cost $4-5 million at least.
?We keep building the resiliency in our network and our NGN project is in the same direction. We are building a new cable, which will connect Egypt to France. So, Flag will have a fully redundant and resilient network,? says Punit Garg, CEO, Flag Telecom.
Bharti also claims to have reacted immediately. ?For certain customers with critical business applications on dedicated circuits, we restored the links over pre-planned restoration routes. For internet traffic, which by its nature can be re-routed around network faults, we immediately diverted the customer traffic towards the Pacific route. Then, we worked with other carriers to augment the Pacific capacity to reduce the level of congestion,? details David Nishball, president, Enterprise Services, Bharti Airtel.
BPOs were also reassessing how they coped. While vCustomer refrained from commenting on the issue, IBM Daksh said it had network redundancy built in and there was no impact. ?There is no disruption of operations or services to our customers as a result of a major undersea cable cut near Alexandria due to the diversity in cable routes and service providers built into our network,? says a Genpact spokesperson.
?It wasn?t a disaster, but a problem that should be addressed. Internet service providers and their customers incurred huge losses. If a BPO had 100MB capacity, it was using only 40MB. We must take a lesson from such incidents,? says Internet Service Providers Association of India president Rajesh Chharia.
As India is connected to two oceans, with the capacity equally divided on the Western coast and the Eastern coast, it managed to put on a brave face, whereas countries like Egypt and Saudi Arabia, struggled to restore connectivity. About 60-70% bandwidth of the Atlantic Ocean, which was affected due to the disruptions, was shifted to the Pacific side for those 10 days. Though this allowed people access to the internet, connection was a bit slow.
?Many areas of the world are connected by a relatively small number of internet cables, and when several failures happen simul- taneously, either by accident or deliberate attack, it seriously affects the ability to conduct business. These events should remind IT and business managers that they always need to prepare for catastrophes with service redundancy and business resumption plans,? says Roberta J Witty, VP-research, Gartner.
Some traffic in these circumstances can be rerouted, but that?s not always the best bet, according to Witty. Satellite backup can be an alternative but it is expensive, and performance may not be good enough for applications that need near-real-time responses. About 90% of internet traffic is routed through undersea cables, and only 10% through satellites. In case of damage to undersea cables, analysts suggest picking up telecom operators that have redundant undersea routes to avoid using satellite as much as possible. Needless to say, internet and phone traffic has to be limited to mission-critical business only.
Typically, cables are laid in proximity to each other and an accident in one can affect others. To cut costs, some cable-laying companies skip expensive marine survey, which could be fatal. ?The best way is to avoid the fault prone routes for which a detailed marine route survey is required before installation of cable,? says Narang. He also suggests tech platforms like multi protocol label switching (MPLS) that make re-routing in a crisis easy. ?An MPLS network is extremely resilient with a core design architecture that has an ability to re-route calls to diverse networks with minimal downtime and high levels of redundancy,? he explains.
The damage to cables had led to a cold war between RCom and Bharti Airtel which is part of the consortium operating SEAMEWE-4. While RCom reportedly said Bharti was overcharging up to 12 times for providing additional capacity for 15 days on its i2i cable network, Bharti dismissed the allegations. Both Bharti and VSNL had re-routed the traffic to Pacific side through i2i SEMEWE cables, respectively.
Also, more cables are being strewn across ocean floors. Alcatel-Lucent and a consortium of nine operators, including Bharti Airtel and VSNL, are also deploying a 13,000-km submarine cable network linking Mumbai to Marseilles in France. India-Middle East-Western Europe (IMEWE) will meet the need for enhanced, backhaul communications capabilities along the India- Western Europe route, as well as for increased capacity to support high-bandwidth broadband services.
Industry experts point out carriers should procure capacities from more than one system specially having cable running through different routes. They should also keep in place restoration arrangements among themselves in such events. ?Bandwidth is a perishable item and should be utilised properly. Rather than one cable network on one route, there should be multiple cables so that services are not affected in case of damage to undersea cables,? says Chharia.
With more than 50 repairs a year in the Atlantic alone, cables breaks are not new to the world. Cables can be broken by fishing trawlers, anchoring, undersea avalanches, shark bites, or even by enemy forces.
Our communication pipelines need to be running 24×7 to keep the BPO engines running. Murmurs of India?s reliability as IT hub were doing the rounds on internet as well as western media. But the country woke up to the significance of undersea cables. While private operators are planning the new cables, the government is asking bandwidth providers to have back-up plans. ?One can?t foresee disaster. So we should have a remedy. Reliance, Bharti and VSNL should come up with a white paper stating what redundancies and back-up plans they are making,? Chharia sums up the mood.