Lower latency, higher security?

22 November 2011 | John Hibbard

The perpetual drive for diverse routes and low latency are enough to keep the industry on its toes.

I was recently at a conference where the convergence of lower latency and network security was raised. Normally these topics are divergent, as added network security usually comes with the penalty of a longer route. This resulted from discussion, both in the conference and the corridors, of some non-mainstream, arguably left-field ideas.

As I mentioned in a previous column, a new route across the Arctic region could offer lower latency so I was surprised to see the degree of planning which had been done recently for both the White Sea and North West Passage routes between Asia and Europe. The melting of the polar ice-cap is certainly fuelling increased entrepreneurial vigour, and we should watch with interest to see how these routes evolve, along with the combination of extra diversity and lower latency they offer.

Just as interesting were the ideas to avoid two of the global “hot-spots” – Egypt and the Luzon Strait. With the former, there was debate over the relative merits of a new route around South Africa versus a terrestrial route across Asia Minor with projects such as LION. Both had upsides and downsides which were interesting to reflect on.

To avoid the Luzon Strait, one idea was to interconnect two recently announced cables – ASC from Singapore to Perth (with terrestrial to Sydney) and Pacific Fibre, to create a Singapore-USA alternative route. Although latency would increase significantly, it would still be less then routing via South Africa.

This concern about latency prompted the suggestion of a scheme to achieve diversity with less latency, which envisaged a consortium of carriers building a cable around the south of the Phillipines and onto the US. This might emanate from Singapore, Malaysia or Indonesia.

Latency would not be significantly different to the current Luzon Strait route, and in Indonesia’s case could be superior compared to routing through Singapore. There was general alignment by those with capacity through the Luzon Strait that they couldn’t afford not to be in such a cable.

Imagine if there was another bust in the Strait which took out most (or all?) of the cables there. There is around 3Tbps of traffic going through there at present. This could not be accommodated on any surviving cables, nor those across the Indian Ocean. Diverse routes via Australia or the southern Philippines would be needed. It will be interesting to see how these developments evolve.

John Hibbard is CEO of Hibbard Consulting. He can be contact at: jhibbard@bigpond.com