Innovations Report, an interesting German site with articles on emerging technologies, had a story last week on some exciting research being conducted right here in Canada (at Carleton University and the University of Toronto).  In an article published on Nano Letters, a scientific journal, University of Toronto Professor Ted Sargent discussed his work on using a laser beam to direct another with unprecendented control - an accomplishment that is quite handy for fiber-optic networks.  Fiber-optic networks can send data at 10 Gbps or 40 Gbps and are being deployed worldwide to connect all the networks.  However, when information is passed from one network to another, switches and routers can introduce delays, slowing down the effective rate of transmission.  The latency is caused by the conversions of data from optical to electronic data, which is in turn used by the switches and routers. 

Sargent claims that the speed of optical (i.e. fiber-optic) links simply cannot be matched by routers.  Going through 10-15 routers in a typicaly hopping sequence does introduce a lot of delay.  The solution was to refine optical switches so that those could forward data at up to 100 times the bit rate of the fastest networks available today.

The joint research project with Carleton Univeristy (Professor Wayne Wang and colleague Connie Kuang) eventually led to the creation of a new polymer material that could be integrated into optical switches.  The end result was a material that combined nanometre-sized spherical particles referred to as "buckyballs" (molecules of carbon atoms resembling soccer balls) with a designed class of polymer.  That combination of the polymer and the buckballs created a thin film with electron-rich molecules with enough power to make light that passes through to control the direction of other light.  That is what yields the switching capability.

This is definitely an interesting development, and pontentially disruptive to the router industry.  Once all-optical devices become the de-facto standard, the delays caused by electronics will no longer be performance bottlenecks.  When can this happen and what are the implications to router vendors?  Well, not quite yet - it is certainly a futuristic technology, because the production of the material needs to be perfected and that same material needs to be able to withstand the tough conditions of the network environment.  Another potential factor that could curtail the uptake of such a solution is the slow uptake of 40 Gb networks.

Note/Update: it might take as much as a decade for this technology to be commercially available, and, more importantly, eventually deployed by some enterprises. But this development certainly represents another milestone for the promise of nanotech.