Aardvark Daily

Please visit the sponsor!

A stroke of genuis?

Modern datacentres chew through an incredible amount of power.

On doing a little research, I discovered that it is estimated the total power consumed by such services is around two percent of the planet's total electricity production. One figure I saw quoted was 90 billion kilowatt hours. That figure is estimated to increase to 8 percent of the world's electricity production by 2030.

Unfortunately, datacentres don't run with 100 percent efficiency so a great deal of that electricity is simply converted to heat and getting rid of that heat becomes an additional cost, in both energy and money.

In fact, cooling is such an issue that Microsoft has even built underwater datacentres as a way of mitigating cooling issues.

Given that a great deal of the planet's electricity is still produced from fossil fuels such as coal or natural gas, thus contributing to climate change, anything that can improve efficiency will obviously be welcomed with open arms.

Well an innovative UK company seems to have come up with an idea that is sheer brilliance, if it works.

The company, called Heata, is proposing to use its datacentre inefficiencies to heat people's hot water and even pay them for the privilege.

This is very much "outside the square" thinking but it has all the elements of genius.

Acknowledging that modern computer technology will always have inefficiencies that create heat, what better way to deal with that heat than deliver it right where it can be used?

Not only do you solve your datacentre cooling problem but you also solve a secondary problem of high energy costs for those who use electricity to heat their water.

I guess that from the datacentre's perspective this also adds a degree of geographical redundancy by distributing its computing resources over a potentially very wide area.

No doubt there will be security concerns, having that valuable data being exported outside of a physically secure environment but hard encryption and other strategies should mitigate that to an acceptable degree.

Then there's the issue of maintenance, both preventative and remedial.

When all your gear is located in one large facility, monitoring, maintaining and repairing it is pretty straight forward. However, if you move that equipment out into thousands (or more) residential or commercial locations you're suddenly incurring a huge extra cost and logistics burden to keep it all running smoothly.

The capital cost of building such a distributed datacentre is also an issue. It's probably a lot cheaper to create a traditional processing facility than it would be to have a huge number of bespoke processors made then installed in a huge variety of dwellings. From this perspective I expect that Heata is relying on significant levels of government subsidy to offset that initial startup cost, if the trial is successful.

Although this seems to be a fantastic idea, "there's many a slip between the cup and the lip" as they say. However, it's this kind of innovative thinking that will probably be essential if we're to end up with a future that is sustainable in the longer term.

Perhaps the home of the future will have a datacentre processor as a core part of its energy system, just like the EV will become the mobile battery and supplementary energy transport system for that home.

Interesting times ahead!

Carpe Diem folks!

Please visit the sponsor!

Have your say in the Aardvark Forums.

PERMALINK to this column

Rank This Aardvark Page

Change Font