At the Data Center Dynamics (DCD) conference in San Francisco, in the US, the following is reported: “At this year’s Data Center Dynamics conference . . . speaker after speaker took to the stage to lament how utility companies refuse to provide even the largest corporations with the renewable energy they want.”

And further: “‘We’re the largest corporate buyer of renewable energy in history,’ Google’s senior director of data center energy and location strategy, Gary Demasi, told the DCD conference last week. ‘And, frankly, it’s very frustrating that we still can’t buy the product that we want to buy and that everybody can’t buy the product that they want to buy.’”

Now, this is very confusing. Let me explain: let us imagine that there is a mountain stream which flows into a river. Once the water from the stream enters the river, it cannot be removed, since it is mixed with the river water. If you want to drink from the mountain stream and only the river is nearby, then you have to drink river water. The only way to drink from the mountain stream is to drink from it before it enters the river.

In a similar way, unless there is a direct connection from the renewable-energy generator to the data centre, the renewable energy is generated into a grid and the grid supplies the data centre. And, ipso facto (which is Latin for ‘fact’, bro) the grid has many generators of nonrenewable energy connected to it.

Now, if the grid was 100% supplied by renewable energy, it would be really wonderful and peaceful, and calm would reign on the planet and the birds (those that have not been sliced by wind turbines) would sing and all races would live together in peace and harmony. But the solar renewables do not work at night and the wind power renewables do not work in still conditions. So, this is the thing: the data centres can only have reliable 100% renewable power if they are directly connected to the grid.

So, what do they expect? For the grid to be 100% renewable power, which is not possible. For the connection to them to be directly connected to a renewable source, which is not reliable? In passing, nota bene (which is Latin for ‘strue’, bro) reliability is not a problem for data centres, since they have huge standby diesel generators, normally 200% oversized, which run a low load, are noisy and pollute like mad. It seems to me that the situation is one of fanaticism, which is defined as “redoubling one’s effort while one’s aim is forgotten”.

And then I thought of project Ivy Mike. As all Natal graduates know, Ivy Mike was the code name given to the first test of a large thermonuclear device in which part of the explosion comes from nuclear fusion. Owing to its physical size and fusion fuel type (cryogenic liquid deuterium), Mike was not suitable for use as a deliverable weapon (it was the size of a three-storey building) and was intended as an extremely conservative proof-of-concept experiment to validate the concepts used for multimegaton fission fuelled detonations.

The idea was: no matter the physical size, build and assemble all the components and see if they will blow up in a satisfactory way. Then use engineering design to make something much smaller that can be delivered by an aircraft or rocket. Thus, using this idea, I thought: is it possible, with an unlimited budget, to create a renewable-energy transmission system that will allow renewable energy to be fed into the grid at one end and to be supplied to a consumer at the other without building a dedicated grid for renewable energy alone?

The exciting news is that I think I have worked out how to do it. Very similar to project Ivy Mike, it will be costly to do a proof-of-concept design and, initially, it will be awkward. I am sure my idea will work. So, read this column next week.