Replacing electrical transmission lines with optical networks

One of the challenges of delivering renewable energy such as wind power or solar systems is the high cost of the electrical transmission lines to carry the power to where it is needed. Unfortunately ideal solar and wind power sites are rarely located near major urban centers. Most renewable energy systems produce relatively small amounts of power compared to a coal and nuclear power plants, and as consequence the cost of the electrical transmission line given the distances to reach renewable energy sites can completely undermine the business case for deploying a renewable energy system in the first place.

But maybe there is another solution of rather than building expensive electrical transmission lines to link these remote renewable energy sites to the electrical grid we instead move our cyber-infrastructure servers, storage and other facilities to the renewable energy sites themselves and link them with optical networks to the global information grid - the Internet.

One of the fastest growing energy consuming sectors is information communication technologies (ICT). It is estimated that ICT consumer upwards of 9% of all the energy output in North America through direct electrical consumption and cooling. Cyber-infrastructure facilities, corporate server farms, etc are major sinks for electrical power and cooling and are putting enormous strains on the electrical systems of our cities, universities and businesses.

With today's modern telecommunication facilities, there is no reason why these cyber-infrastructure facilities and server farms need to be located in close proximity with their users. High speed optical networks allow these facilities to be located anywhere. In fact many large corporations like Google, Microsoft, Amazon and others are already starting to collocate their server farms to low cost energy sites around the world.

The obvious next step in this evolution is to collocate cyber-infrastructure equipment and servers directly to the renewable energy sites themselves. And rather than building expensive electrical transmission systems to connect these renewable energy sites to the electrical grid, we instead build much cheaper optical networks to the servers to interconnect them to the global information grid - the Internet.

One downside of this approach, is that these cyber-infrastructure facilities and servers will not be connected to any electrical grid, and as result they will experience a lot more outages and down time dues to the waxing and waning of the wind or the diurnal cycle of the sun. But the beauty of ICT is that we already have the technology to do rapid load balancing of servers due to outages, and of course, the Internet from day one has been designed to route around outages.

We have the technology at hand to build "follow the wind" or "follow the sun" computing grids using optical networks to ensure extreme high reliability information systems and computing grids regardless of whether or not components of the underlying physical computational network and/or storage facilities are available and on line. The mesh of global optical networks around the world will further help provide load balancing due to varying wind and solar conditions.

Ben Bacque of Alcatel-Lucent has even suggested that we locate these renewable energy/server farms in Canada's remote artic regions because this would also help address the cooling challenges of todays modern servers. Up to now it has been impractical to locate renewable systems in Canada's high north because of the high cost of building transmission lines over immense distances across inhospitable terrain.

Building optical networks to remote renewable energy systems will also allow governments to achieve an important social objective of delivering high speed Internet to remote and rural communities and would provide much needed jobs for the maintenance and care of these server farms and renewable energy systems.

Optical networks can be also used to interconnect micro-power systems that provide power to peer to peer storage and computing grids. As with renewable energy systems, the existing electrical grid is ill suited for connecting hundreds, if not thousands of small micro electrical power systems located at our homes and businesses. The interconnection to the grid requires costly and expensive switches and meters that must be installed by professional electricians, and the distribution system must be re-configured to handle power origination from those who were traditionally consumers of electricity.

So rather than connecting the micro-power systems to the electrical grid we can perhaps use them to power locally hosted servers and storage facilities. And as before these servers and storage facilities can be interconnected via a well proven peer to peer grid over the Internet.

Robin Chase sent me an interesting pointer to a talk given by John Holdren (Director of Woods Hole Research Center) to the UN in September 2007, John Holdren's slides had a stunning number: If worldwide CO2 emissions peak in 2015 – that’s seven years from now – we have a 50 percent chance of avoiding catastrophic effects of climate change.

http://networkmusings.blogspot.com/2007/10/closing-climate-change-window-of.html

Most scientists already think we are at a tipping point in terms of CO2 emissions.

To my mind the ICT industry and research community as a whole has a moral responsibility to help address this problem. If the ICT industry and research community consumes 9% of the global energy budget, we can safely assume that ICT contributes to 9% of the worlds global emissions of carbon dioxide. But as opposed to any other sector in society ICT community has the means and tools to virtually eliminate this entire carbon footprint (and possibly more) through the many techniques I have outlined in this blog and previous postings. A 9% drop in carbon emissions over the next decade would dramatically mitigate the threat of global warming.

We need a call to action by the ICT industry and research community. We need to start immediately testing and experimenting with these ideas and many more that I am sure will be thought of in the process of identifying possible solutions. We need to immediately freeze the carbon budgets of our universities, research centers and server farms. Universities and research centers are the institutions that should be demonstrating global leadership and developing new solutions to address global warming.