Polyfiller or Polycrisis??

Another day, another new buzz word – maybe or not.

What on earth is a polycrisis though? The Cambridge Dictionary defines a polycrisis as “a time of great disagreement, confusion, or suffering that is caused by many different problems happening at the same time so that they together have a very big effect”. Concurrent shocks and interconnected risks combine to create a crisis even worse that the sum of it’s parts.

Does our industry actually face such a scenario? If you have been living under a rock for the last couple of years, maybe you don’t see the scale of the problem we are facing. However, those of us who live in the real world, may have heard industry commentators talking about the problems facing the industry such as, but not limited to:

1.????? AI energy usage

2.????? Water usage

3.????? Public perception

4.????? Carbon Usage

5.????? Cooling

6.????? Utility usage and cost

7.????? Renewable energy

8.????? AHJ Permissions

9.????? Governance & Regulations

10.? Space Availability

11.? Lack of talent

12.? Supply chain challenges

13.? Growth of Cloud

So out of the 13 challenges listed above nearly all of them affect every data centre, existing or being built. These are just the headlines, sitting under them are a whole raft of issues to be dealt with. For example, cooling could look at air, chilled water, liquid cooling, re-use of heat, environmental conditions of data halls and the external environment which has rising ambient temperatures.

We cannot just focus on one or two areas as all of these issues are interlinked and have interdependencies.

Far more intelligent people than me have spoken about the individual areas of concern, however, I now see as a crisis worse than the sum of its parts; a polycrisis!

We continue to see predictions of growth and how new GPU’s and the like will need additional power and cooling. However, the designs we have used up to this point require a radical re-think. Continuing to do the same thing time after time and expecting different results is the definition of madness. We need to change and change now.

Advances are being made in design; however, this needs to be done while looking at the whole lifecycle of the proposed solution. For example, look at wind turbines. They have a life of between 20 and 25 years before end of life. Most of the turbine, shaft etc can be recycled, except for the 60-80 metre blades. As these are made of a composite material it means that they can be very difficult to recycle. Besides to cut them into smaller lengths for easier transportation is also not an easy task. One manufacturer of turbine blades says that they have discovered a new chemical process to chemically ‘disassemble’ the blades. However, the website https://decomblades.dk/ has 3 recycling technologies to be used; Pre-processing and Shredding, Cement Co-Processing and Pyrolysis. The latter two methods are energy intensive, the amount of energy in cement is well known. Pyrolysis is still at the ‘science project’ stage and is not available at scale. Was this thought through with the original design of wind turbines, it would appear that it wasn’t. ‘It is happening in 20-25yrs time hopefully a solution will have been reached by that stage’. However, here we are, with 45,000 blades due for end of life. The sad but inevitable solution is to landfill them as per the picture below. I call this situation out as a case in point that, as a species, we are not particularly clever in thinking things through to the logical conclusion.

A good example of this is that reported on by https://www.datacenterdynamics.com/en/analysis/how-to-build-when-the-power-isnt-there/ While the use of fuel cells is well known, what is not so common is to understand the by-products of the fuel cells H20 and C02. Therefore, if you build an urban farm next-door this can make use of these by-products. By purification the C02 can then be provided at a food grade and sold off. The fuel cells are of course hydrogen ready. Also, the facility becomes water positive as well. This is a project that has been thought through well as it helps to mitigate some of the carbon emissions of the natural gas which fire the Fuel Cells. This also uses a cheaper fuel source than the grid electric power. Although this project is currently a proof of concept (science project) it does help to establish what can be done.

I hope that the above demonstrates the need to consider design as a whole and not elements in isolation. If we, as a community, fail to do this we are not taking these issues seriously.? We need to do better and need to do it now!