Supercomputers: Technology With the Power to Solve the World’s Most Complex Problems

Supercomputers have been around for a while–since the 1960s– and yet, they remain, in many ways, the technology of the future. As they become more prevalent and advanced, they will help solve some of the world’s most complex problems. In fact, they already are. It was supercomputing that helped create the Covid-19 vaccine by aiding in the identification of the virus’ spike protein. 

Computer scientist Jack Dongarra explains the nearly unfathomable power of supercomputers, saying “If everybody on Earth were to do one calculation per second, it would take four years to equal what that computer can do in one second.” 

A recent breakthrough was made when a supercomputer broke the exascale barrier for the first time. The supercomputer, known as Frontier, is now the world’s fastest supercomputer. Breaking the exascale barrier means that supercomputers are now able to solve calculations 10x as fast and at 8x the complexity. Frontier employs approximately 50,000 processors, which compares to the most powerful laptop that comparably uses 16 or 24 processors. 

Due to the astronomical number of processors, these supercomputers are expensive to make, coming in at around $600 million, and are energy-expensive to run, utilizing 20 million watts, as compared to a normal computer’s 65 watts. Over the course of a year, these supercomputers can use as much power as a small city. 

These computers also take up quite a bit of space. Depending on how many computers make up a given supercomputer, they can vary in size. Generally, supercomputers take up more than 10,000 square feet. For obvious reasons, these computers need to have access to high amounts of affordable electricity and need to be stored in a cool environment deep underground, to avoid radiation or cosmic rays. 

What’s the Difference? Quantum vs Super-Computing

Supercomputers have been around since the 1960s, and due to their immense size and energy demands, their ability to expand is limited by these factors. Relatedly, quantum computing was pioneered in the 1980s, and instead of relying on binary code, like computers and supercomputers, they use small information-storing units, known as qubits.  

Despite differences in how these two models of computers store and process data, both computers are exceedingly competent. Back in 2023, an IBM quantum computer beat a supercomputer on a benchmark test. Quantum computers are thought to be faster at problem-solving than conventional supercomputers. 

With that in mind, each type of computer has its best applications, and some speculate that in the future, the distinction between supercomputers and quantum computers may become less distinct, if that distinction persists at all.

Supercomputers and AI

Despite the size and energy demands of these computers, we are sure to be seeing more of them, especially since these supercomputers can power AI models. AI is increasingly dependent on supercomputers for research and development. Many major tech companies, among them Meta, Tesla, and Microsoft, are working on developing their own supercomputers. 

AI and supercomputers, used in tandem, would complement each other and further research in key areas such as science and medicine, and sustainability. It is speculated that supercomputers will be able to make developments in drug development, for example, by cutting back on the discovery time, or could help develop more sustainable alternatives to elements of our everyday lives. 

While supercomputers have been around for a while, and while they may lack the flashiness of market-ready AI, they are undoubtedly behind many of today’s innovations and will continue to power tomorrow’s discoveries. 

Stay up to date on our latest posts by subscribing to our newsletter and following us on LinkedIn and Instagram.