Sean Park Portrait
Quote of The Day Title
There's no bad time to innovate.
- Jeff Bezos

Saving Iceland(?) – ICT meets Climate Change meets Markets.

Flying Around Iceland on Google Earth in iPhone
Image by cogdogblog via Flickr

A pretty interesting space I’d say. I’d love to be able to attend the Greening of the Internet conference in a couple weeks but unfortunately it would be a bit of a luxury now and logistically impossible as well so I’ll just have to follow remotely. (If anyone spots a good live blog or two on the conference please let me know in the comments…)

Over the past couple years, I’ve become increasingly interested in the potential to combine technology, energy and markets to solve multiple problems concurrently. One of the absolute best people to follow if you are interested in this space is Bill St Arnaud; I highly recommend subscribing to his RSS feed or email list.

One back-of-the-envelope idea I’ve been touting for the past few years is that Iceland should reinvent itself as (one of) the data-centers to the world. Given the complete implosion of the Icelandic economy in 2008, if my idea makes any sense, it should be pursued even more aggressively. In a nutshell…

As the world inexorably moves to computing and storage as a utility, and as energy prices and climate change (price of GHG emissions) loom ever larger in the economics of cloud computing and data-centers, and as information transmission (bandwidth) costs continue to plummet, the relative importance of (carbon-neutral) energy operating costs will continue to increase strongly. (The other main factors being: construction costs, access to bandwidth, legal jurisdiction, security, availability of skilled workforce.) Sooo…. why Iceland? Cold climate and abundant green energy (geothermal now, offshore wave and wind later.) Heck the forward stream of CER’s (depending on what happens in Copenhagen) might even fund the upfront capital cost! Also one would guess that Iceland (as opposed to say Greenland or the Canadian Arctic) would have both reasonable construction costs and access to the skilled workforce needed to both build and then run these operations. I’m not an expert on data-centers but I don’t think they generate massive amount of jobs, but if this vision were to become true, Iceland could build a whole ecosystem around Green ICT and given it’s small population, I’m sure this would have a material impact on both employment and GDP over both the short and the long term.  the= The biggest risk I can see would be natural disaster risk (ie volcanoes blowing up your data-center) and risk of damage to sub-oceanic fibre going in and out of Iceland. Both of which I think should be reasonably easy to mitigate and insure against. Plus, on the face of it, its geographic isolation and NATO membership probably make it a good choice in terms of security.

So…if I were managing a substantial Private Equity Infrastructure fund, I would (do a proper analysis of course but assuming my back of the envelope thinking holds):

  • Start talking to the Icelanding government, IMF and the World Bank/EIB about co-funding and incentive schemes (especially the required incremental undersea bandwidth
  • Encourage Iceland to join the EU asap because having data-centers in a EU jurisdiction will be a very important commercial imperative.
  • Start talking to Amazon (AWS) and Google about partnering / risk sharing as potential ‘anchor tenants’ (they could become to Iceland what Dell was to Ireland 20 years ago…)
  • Start talking to smart banks, hedge funds, etc. about innovative carbon-based financing and energy hedging structures.

Obviously this is just scratching the surface of the work that would need to be done but equally the myriad of collateral opportunities that would emerge as a part of such an ambitious project.

I’ve decribed this idea with varying degrees of detail to probably 2 or 3 dozen people over the past 2 years and no one has found a serious or obvious flaw in the logic. Then again none of them – myself included – had “done the math” nor was necessarily highly fluent in the mechanics/costs of building data-centers generally. Equally, none of them turned around and offered to raise a few million of seed finance either to do a feasability study, so you could say talk is cheap. So in the spirit of casting the net wider, I thought I’d write about it here and solicit experts to punch as many holes in the idea as it merits.

In any event, I never understood why Iceland put all its chips on banking and retail when it is so clearly at a comparative disadvantage in both domains…if nothing else my scheme at least has the semblance of logic behind it! ;)

Reblog this post [with Zemanta]

Post to Twitter Tweet This Post

  • I was somewhat disappointed not to have anyone comment on this proposal...perhaps no one saw it? wrong audience? I want someone to tell me why this isn't a brilliant idea.

    Recently came across this paper - thanks to the always interesting Bill St. Arnaud - from academics at Rutgers, which puts a bit of 'meat on the bone' so to speak.
  • And maybe the heat generated by all those data centres could be used to warm the rest of the country during winter (or conversly, the low temperatures of Iceland might mean less energy spent on cooling the data centres in the first place)
  • Exactly. Indeed all data centers should be built as far north (or south) as possible to harness the natural synergy with a cold climate. In a nutshell, data centre location should optimize:
    * cold prevailing climate
    * proximity to reliable renewable energy (geothermal, wind, wave, solar...)
    * cost of connecting to the network

    Obviously an oversimplification, but if externalities were correctly priced you would never build a data center in Arizona for example...
blog comments powered by Disqus