India Averts Internet Meltdown

Failsafe plans, including an alternate network of cables in the Pacific Ocean, prove the resilience of the country's broadband infrastructure

When news broke Jan. 31 that an undersea fiber-optic cable owned by India's Reliance Communications had been accidentally sliced by a ship's anchor in Egypt, people feared a possible replay of 2006. That's when an earthquake in Taiwan disrupted Internet traffic in East Asia for nearly two weeks. India, where the Internet is the lifeblood of the outsourcing industry, was considered particularly vulnerable.

Happily for Reliance (which did not respond to phone calls), and for Indian outsourcing giants Tata Consultancy Services (TCS) , Infosys (INFY), Wipro (WIT), and Satyam Computer Services, which use such undersea fiber-optic cables, no serious problems occurred. Neither, it appears, have the global operations of multinational tech players like IBM (IBM) skipped a beat. True, some of India's Internet cafes saw connection speeds slow dramatically, so the time needed to connect to Google increased 60 times—from two seconds to two minutes—but overall the economic impact of the accident has proven minimal for India.

It could have been much, much worse. According to the Internet Service Providers Assn. of India, nearly half of the 25 gigabits of bandwidth India uses relies on cables, of which 60% snake under the Atlantic Ocean. The rest is routed through the Pacific.

Internet Traffic Priority System

Instead, the disruptions gave India's outsourcers the opportunity to prove to their clients how well they're prepared for such emergencies. As soon as the problem with the Reliance cable became known, traffic was rerouted through the alternative network of cables running under the Pacific Ocean. "There are redundancies built into the system. There are multiple pipes, so if one breaks, others work," says Pradipta Bagchi, a spokesman for Tata Consultancy Services.

Bagchi explains the pecking order for Internet traffic priority. Top of the heap is voice—so that calls to Aunt Mathilda in Missouri or cousin Vijay in Bangalore are unaffected. Then come international private leased lines—networks that banks such as JPMorgan (JPM) or tech service providers such as IBM (IBM), TCS, and Infosys use to connect to clients and their other locations across the world. "These links get high priority and get switched automatically to another line," Bagchi says.

For this privilege, the larger companies pay double or triple the cost to telecom providers such as Reliance, AT&T (ATT) , VSNL, and Bharti Airtel. They also have contracts with multiple providers, so if one fails, the other kicks in. So on Jan. 31, as Reliance announced its problems, rival providers picked up the traffic.

Operations Functioning Seamlessly

However, smaller back-office outsourcers probably didn't fare as well during the Internet outage. Because these companies compete on price, they are unlikely to be able to afford the kind of contingency plans TCS has with Internet service providers. But even their problems are fairly limited. "Though there is a marginally slower access to Internet, this has not disrupted or affected our operations. Our IT team is working closely with our service providers to ensure that operations continue to function seamlessly," says P.V. Kannan, founder and chief executive of 24/7 Customer, a smaller Bangalore outsourcer.

A day later, says Rajesh Chharia, president of the Internet Service Providers Assn. of India, almost 80% of the bandwidth has been restored, as companies made the automatic shift to alternate providers. "The service providers have released more capacity on the Pacific route," he says.

Before it's here, it's on the Bloomberg Terminal.