How Supercomputers Are Affecting the Environment

How Supercomputers Are Affecting the Environment

People are creating more powerful technology every day. But have we forgotten something along the way?

Supercomputers may be our glimpse into the future. Renowned computer scientist John von Neumann proposed the term technological singularity to describe the point of development at which it is impossible to predict the changes that technology will bring to human civilization. We are undoubtedly moving towards that singularity, and supercomputers may be the first giant leap for humanity. So, what are supercomputers? Basically, they are computers, but way better than the one sitting in your room. They have more than one CPU, which allows them to achieve better performance. 

While they can provide remarkable calculation power, the amount of damage they have done to the environment makes one ponder—are they truly helping humanity? Every day, there is news about how climate change is getting worse, and we see reports on the effects of coal plants or vehicles on a daily basis. However, the ecological impacts of supercomputers remain unnoticed, despite the role they play in worsening the climate crisis. In this blog post, we will learn how supercomputers are affecting the environment and how we may fix them before it is too late. Stay tuned! 

How superior are the supercomputers?

Supercomputers are the kings of the digital world. Their superior computational power brings a lot of benefits to humanity. For example, American officials use a supercomputer in the Stockpile Stewardship program which is related to nuclear weapons’ safety. Specifically, the officials would conduct simulations that require enormous computational power, which can be provided by the supercomputer, to ensure the nuclear weapons are safe in storage. 

The Japanese representative takes the crown as the fastest supercomputer in the world. As the king of all supercomputers, the Fugaku can achieve 1.42 exaFLOPS. “FLOPS” is the acronym for floating-point operations per second. You may interpret each floating-point operation as a step the computer runs. Still, the steps involve incredibly complex numbers, making the calculations harder than the ones regular computers handle. So, in other words, the machine can solve 1,420,000,000,000,000,000 very complicated calculations per second—no wonder the machine can handle sophisticated tasks, like weather forecasting. 

The computer world is pretty simple. For more complicated tasks, you will need more power. Supercomputers are all monsters when it comes to power consumption. For example, consuming an enormous power of 29,899 kW/h, the Fugaku has a power consumption of 175,000 traditional desktop computers

The dark side behind their superiority

While supercomputers are contributing a lot to humanity with their power, we must also acknowledge the harm they bring to Earth. For example, there will be more water needed for cooling the hardware with such stunning computational power. As a reference, a data center with a power usage of 15MW can use up to 360,000 gallons of water per day. In comparison, the Fugaku has a power consumption of 30,000 MW, which requires 700,000,000 gallons of water for cooling. 

Besides, their energy consumption is also concerning. A 1,000 petaFLOP computer, or one Fugaku, would take an entire coal plant to power. What’s more? Nvidia has announced a new supercomputer in production called Eos, which is expected to provide 18.4 exaFLOPS of artificial intelligence (AI) computing performance—that will require the power of 18 coal plants. The damage such machines bring to the environment is immeasurable. 

Ways we can reduce the amount of energy used by supercomputers 

While the entire industry seems to be focusing solely on computational power, the Green500 is established. It provides a list of supercomputers that are highly energy-efficient, serving as a warning to the industry that we should pay attention to the environment. 

The founder of Green500, Wu-chun Feng, is trying to utilize the CPU better. He describes it as “mapping the right task to the right processor at the right time.” They believe that this way, it can achieve better energy efficiency. To do this, Feng proposed that supercomputers should use specific, finely-tuned CPUs for supercomputing. 

There are also some unexpected ways to lower energy consumption. For instance, placing the units closer together. This may also reduce the energy consumption, as the data needs to travel a lesser distance now. Well, one centimeter is already a significant distance in the computer hardware world.

Why reduce the amount of energy used by supercomputers?

Supercomputers are at the forefront of our technological development. There is no doubt that machines 100 times the power of the Fugaku will appear in the future. That is part of our development and is our journey to reach the prophetic technological singularity.” But when that happens, if the energy efficiency remains as bad as it is, how much harm would it cause to the environment? Supercomputers represent not only the peak of human intelligence but also the future. We should deal with the problem once we discover it, but not wait until it gets severe and irreversible. 

Perhaps it is time we take a break and reflect on our development. Power consumption is already a severe problem for humanity. And throwing more and more coal plants into it without coming up with any mitigating measures will only worsen the situation. Indeed, we have all these supercomputers, and the technological singularity may already be right before our eyes. But while we indulge in our advanced technology, we also have to make sure that we do not neglect some other core values of humanity——that we will be living in a utopia like Star Trek, not a dystopia like Blade Runner.

Also Read:

Header image courtesy of Pexels

SHARE THIS STORY

Share on facebook
Share on twitter
Share on linkedin
Share on email

RELATED POSTS

How Can Cities Manage Traffic Effectively with AI?

In metropolitan areas around the world, traffic congestion is a nightmare for many. It wastes our time, ramps up our stress and exacerbates pollution. To combat this major headache, there might be a promising solution on the horizon: artificial intelligence (AI). Many urban areas are already turning to AI-powered tools like responsive streetlights and systems that predict traffic patterns to smooth out the bumps in our daily commutes. Given the adaptability of AI, these systems can learn and improve over time, refining their algorithms to meet the distinct needs of different cities.