Renewable energy sources - the good, the bad and the ugly

Headline figures for data centre energy use are grossly misleading.

Traditionally, every business that used computers (pretty much every business) would have their own server room with enough computing power to run the business, plus backups for resilience. The computer equipment in those rooms would itself consume a hefty amount of energy. The redundant backup servers that are there for automatic failover in case the primaries fail would also be on, consuming power. That much power use generates a lot of heat, so there'll be air-conditioning running in there on full whack all the time. Backup media will be travelling daily by road to secure off site storage. Multiply all this across all businesses and the energy demand is enormous.

Nowadays, businesses are increasingly moving their computing to the cloud. That means huge data centres. In doing so they can significantly reduce the computing power, and the associated energy demands, of their own individual server rooms. Yes, the problem is being moved elsewhere, but in a shared data centre, overall there is much greater efficiency. Each business no longer needs it's redundant failover servers. The data centre will have a few shared backups. Each business doesn't even need computing power sufficient for peak demand all the time. In the shared data centre, resources will be shuffled about as necessary to meet demand, so if one business runs all its heavy batch processes at 6 am and another does it at midnight, all that happens is the data centre allocates additional resources to one, takes it back when complete, and allocates to another. If a business leaves all its servers running overnight because one job runs at midnight, it's still using lots of power. In the cloud (in a shared data centre) any resource that is not activity is parked, consuming virtually no energy (like a laptop in sleep mode). Then there's the cooling. It's easier to cool blade servers in a purpose built building than it is to cool kit that is all in its own individual cases inside a room that was originally designed as a storage room.

The granularity of resource sharing in a data center is much greater than the equivalent number of individual server rooms, and the dedicated data centre will be designed and built with efficiency in mind. The operators are running a business, so don't want to be paying for energy that's wasted. The IT manager in the small business is unlikely to get budget for a bunch of top experts to come in and squeeze maximum possible efficiency out of their server room. The company running a data centre absolutely will, because that up front design cost will save them a fortune in the first year alone.

So yes, data centres use huge amounts of power. But that needs to be compared against the power that individual businesses will save by moving their on site computing power to the cloud. In most cases it's a win win all round.

 

Good. It was never going to be as eco friendly as they wanted us to believe. Most of the hydrogen was going to be produced by splitting hydrocarbon, ie fossil fuel. Instead of people burning fossil fuel the plan was to burn some fossil fuel to generate the huge energy needed to split more fossil fuel to produce what they term blue hydrogen.

I visited one of the green washing drop in centres where they had a model suggesting it would be done by electrolysis powered by offshore wind. I asked two questions, which they couldn't answer.

1. Is there enough capacity at the offshore wind farm in your model to meet demand?

2. Given that the offshore wind farm in the model is currently supplying electricity to homes and is not enough to meet all the electricity demand today, how will you generate the electricity to meet current demand once you start diverting that power to hydrogen production?

The poor rep in the end proposed I meet with the director who might be able to explain it. The director was meeting with the public on a given date, but I needed an invite. I didn't get an invite.

 

Not really surprised at this.
Apparently the Government is to announce support for Small Modular Reactors (SMR) as developed and promoted by Rolls Royce, not exactly renewable and you still have the waste problem, but better than burning hydrocarbons. Also quicker and easier toroll out than a full size plant like Sizewell C etc.

 

Good. Hydrogen has for some time been described by government (and its advisory groups) as a niche product. Using it as fuel for gas boilers (apart from delaying a switchover, which I expect is a manufacturer hope) was never a good idea, considering the costs generally.

 

I think the government should actually be partnering in the RR-led initiative. There is export potential, too. I don't think we should buy the project in.

 

Disagree on this the Government should not partner anybody on this; not unless we want the whole project to grind to a halt. The Government should facilitate by getting sites connected to the grid, easing planning permission etc.
Rolls and their partners have brought it this far and this morning said that they could raise the finance from the City.
The technology was originally developed for nuclear submarines by Rolls Royce

 

I kind of agree with Nigel, as soon as the governments get involved its a downhill spiral.
They, whatever colour, only look as far ahead as the next election, they seem unable to plan further than that.
Mostly because they know the next government is likely to just stop anything the government before them put into motion.

 

In an ideal world, I think I'd want the government to invest on behalf of the people, and then make sure the people got a fair deal, and that most of the profits came back to the government to be reinvested for the people.

Take a moment to get the laughing under control, get your breath back....

In the real world of course that's not how it would work. The work would be contracted out to companies that see that it's a government contract and start excitedly rubbing their hands together, knowing they can charge ten times the going rate. The government will ensure good value for the taxpayer by appointing an 'independent' auditors firm like PWC or Delloites, who will also charge a couple of grand per consultant per day to annoy the contractors and slow them down by making them explain why they need to spend two quid on a box of pens. Then if/when it's somewhere near completion, the finished product will be sold to France or China who will leach all the profits out while not actually investing much in maintenance, until the project eventually dies.

 

Sounds like a water company where the so called regulator just took the money and sat on its hands.

Governments have a track record of piling in loads of our money then scrapping them, they are best not involved in anything that matters.

 

From that article:

"A final decision on each project will not occur before 2029 – a date at which Rolls-Royce had hoped to be up and running – with power being produced from the “early 2030s”"

That is very disappointing. We are now years behind expectations. I wonder why precisely this has happened.

I have not followed the various committees since this:

Parliamentlive.tv

Which seemed full of hope.

 

Isn't there problems updating the grid?
You can come up with all kinds of energy solutions but if you can't connect it to the grid its worthless.

I can't understand why Dungeness isn't a prime site for this kind of stuff as all the grid is already there.