You know the submarine cable boom is back when Google is building its own network.
The bandwidth bubble of the late '90s and its subsequent bust around 2001 remains one of the great milestones of the telecoms industry. But seven years, several bankruptcies and numerous M&As later, submarine cable projects are back in style. TeleGeography counts no fewer than 25 new cables scheduled for construction worldwide between now and 2010, at a cost of around $6.4 billion - and those are just the ones that have been green-lighted.
And while many of them are focused on under-served regions like Africa and the Middle East, Asia Pacific is also seeing a number of new cables on trans-Pacific and intra-Asia routes, as well as on routes linking India to the Middle East and Europe. By the time they're completed, TeleGeography says, Asia will have more potential capacity than the fiber-heavy Atlantic.
One of them, the trans-Pacific Unity cable, announced in February, is something of a milestone in itself as Google is one of six investors. The buzz over Google getting in the subsea cable business isn't just about the novelty factor. With the rise of Web 2.0, P2P, streaming video and other bandwidth-intensive apps, a number of internet pundits and experts have been predicting an 'exaflood' of traffic so massive that all the excess capacity built during the boom would be gone in a couple of years. Meanwhile, the fact that a web company needed so much trans-Pacific capacity it decided it would be cheaper to build than lease was arguably a sign of the times that the massive wave of IP traffic that subsea operators had been expecting years ago is finally on its way.
But you wouldn't know it from all the unlit capacity. According to TeleGeography, less than 20% of potential capacity was lit on existing trans-Pacific cables on major subsea routes by the end of 2007. For intra-Asia, that figure was under 10%.
That's not to say that IP traffic isn't on the rise. TeleGeography research director Alan Mauldin says that international internet traffic grew 53% between mid-2007 and mid-2008, 'fueled by consumer demand for video, delivered via web browsers, peer-to-peer services, or streaming protocols.'
But what's really driving the boom, according to Mauldin, is 'a need for a broader range of restoration options in case of cable failures, the desire for more diverse routes between two destinations, and stubbornly high capacity prices in markets that have ample capacity but few competitors.'
All are valid incentives, particularly the restoration angle following the recent rash of major cable breaks in various locations, as well as the Taiwan earthquake in December 2006 that famously severed seven of nine fiber links in the Luzon Strait.
Byron Clatterbuck, SVP of Tata Communications' Global Transmission Services group, adds that many consortium members are driven by the need to control their opex, and owning fiber enables them to lock the cost of capacity.
'They don't want to be subject to the whims of the market,' he says.
But survivors of the earlier bandwidth boom could be forgiven for looking at all the unlit capacity in the water - to say nothing of the current shaky economic climate in general - and thinking, 'Here we go again.'
Intelligent control plane