In Europe this morning, Colt Group (LON:COLT, news) launched an attack on the low latency bandwith marketplace. The trans-european fiber operator is now offering 4.22ms between London and Frankfurt as well as 2.65ms between Paris and Brussels. The former has been a battleground for a while of course, but the latter is a new one for me. Is the Paris-Brussels route catching the low latency financial trading bug too? The more the merrier I guess! It’s refreshing to get actual numbers for the latency on these routes, although it is just a snapshot since they are continually driving further microseconds out of the network over time.
According to the PR, Colt achieved these numbers by ‘building two new, direct fibre routes and then applying its next generation technology to ensure the fastest possible transmission.’ By building, I suspect they don’t mean they dug new conduit the whole way, but rather that they pieced together a better fiber route via various sources across Belgium, which is how I understand the low latency battle has been fought – at least on the London-Frankfurt route. The various European fiber builds of a decade ago were built in two loops that didn’t really keep low latency between London and Frankfurt in mind, so operators have to ‘go offroad’ to compete. I had heard that Level 3 Communications (NYSE:LVLT, news, filings) was on top on that route earlier this summer, but it’s a giant game of leapfrog at the moment.
Enabling Colt’s offering is gear from Infinera, with whom the operator has had a relationship since last year. Infinera has been playing a prominent role in many low latency network offerings, including that of Spread Networks on the NY/Chicago route. I wonder how much it contributed to their Q2 earnings surprise.
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!
Categories: Internet Backbones · Low Latency
Join Our LinkedIn Group
More interesting developments. Paris-Brussels is interesting as most of these routes are fuelled by matching engine to matching engine requirements for trading.
London to Frankfurt is getting more competitive, there is another story about shortest metro routes, to make the most of the low latency long haul piece. A sub-optimal metro route can make an attractive “PoP to PoP” figure look a lot less attractive once that is factored in.
It would be worth getting a look at the numbers from a matching engine in London – LSE, Chi-X etc, to Deutsche Borse, as an end to end measure.