Print

Comment

Email

Digg

Del.icio.us

Reddit

The recent spate of undersea cable failures reportedly cut India's bandwidth by about 40-50%. And raised a lot of questions about the security and reliability of these underwater communications channels.

The first interesting thing was five cable cuts reported within the space of a few days. The damaged cables included the FLAG Europe-Asia cable near Alexandria, the FALCON cable near Bandar Abbas in Iran, the SeaMeWe-4 cable near Alexandria, SeaMeWe-4 near Penang, Malaysia, and FLAG near the Dubai coast. Ideal material for conspiracy theories to be floated. These included an American attempt to cut off Iran from the rest of the world, a precursor to an Islamic terrorist organization's attack on the Western world and NSA installing taps in the cables using the USS Jimmy Carter.

Closer to home one allegation was that this is a targeted attack on India's thriving BPO industry. Thankfully they have been by experts, who say that to have actually caused any significant impact, a lot more would have needed to be done. All that these events have done is inconvenience a lot of people, rather than causing any real damage. Meanwhile, the countries which were affected simply rerouted the traffic through satellite-based networks.

According to TeleGeography Research, cable cuts happen once every three days on an average. Friction against rocks on the sea floor can cause cuts. This is completely normal, and repair ships are always on the lookout for damaged cables to repair. Users don't get to know about such outages because most ISPs don't rely on a single link and are able to reroute the traffic from elsewhere. This time, however, this did not work as well.

Shyam malhotra editor-in-chief VOICE&DATA shyamm@cybermedia.co.in

At least not Reliance Communications, which owns both FLAG Telecom and FALCON. Bharti and Tata, part of a consortium of operators that owns SeaMeWe-4, were able to reroute traffic by taking it across the Pacific instead of the Atlantic. Reliance, unable to find a redundancy option, approached TRAI, requesting that it establish a code asking operators to share capacity with rivals in case of disruptions. Interesting thought. Except that these arrangements need to be discussed before the event. One wonders if any arrangements will be put in place before the next set of such accidents.

What about the users? A survey conducted by FICCI on the 2nd of February among India's IT/ITeS companies revealed that the larger organizations felt little to no impact of the damage to the cables near Alexandria. Most of them have well-defined redundancy plans in place and were able to switch providers to avoid downtime. The hardest hit were the SMEs across the country, who mostly rely on a single ISP. So redundancy now has to be not just about service providers, but also about where the bandwidth is being routed from-land, sea or space. CIOs may now want to pay closer attention to this when shopping for bandwidth. Will that drive up costs? Particular combinations might but overall there should be little impact.

So, can anything be done to fix the undersea cable security? A lot is done-informing fishermen and navies, building cable routes to avoid areas prone to earthquakes-it is near impossible to secure a cable network beyond a certain degree. The landing stations can be secured but how do you protect something from deliberate human damage when it is under a mile of water?

The best answer appears to be redundancy. We need more cables. Many new cable systems are expected to come up in the next few years. These include Telecom Egypt's TE North cable, Orascom's MENA system, the IMEWE cable, and a new cable by FLAG Telecom.

This will definitely reduce chances of failure by providing additional options for the routing of Internet traffic. That seems to be the best answer.