Where big data meets massive computing power—welcome to the age of supercomputers.
Supercomputers are incredibly powerful machines designed to perform complex computations at speeds that standard computers can’t match. These high-performance systems can process billions of calculations per second, making them essential for tasks that require vast amounts of data analysis.
You won’t find these machines in your local electronics store—they’re prohibitively expensive, costing anywhere from millions to billions of dollars to build and maintain! That’s why supercomputers are usually deployed in large-scale projects across fields such as climate research, quantum physics and advanced medical research. Here, they enable scientists and researchers to tackle complex problems and conduct simulations that would be impossible on ordinary computers.
In this article, we’ll explore five pivotal projects where supercomputers are playing a crucial role, dramatically saving time and pushing humanity closer to a new frontier.
Human Brain Project
Launched in 2013, the Human Brain Project (HBP) was a large-scale research initiative set up to advance brain research through digital simulations powered by high-performance computing. Its primary goal was to create a comprehensive digital model of the brain to explore its complex functions. Given that the brain has around 100 billion neurons and 100 trillion connections, this was no small feat—requiring the kind of computing power only supercomputers can provide.
Different layers of a personalized brain network model.
Image from the Human Brain Project
To achieve this ambitious goal, the HBP built EBRAINS, a digital research infrastructure. EBRAINS offers a wealth of brain data, a multi-level brain atlas and advanced tools for modeling and simulation. It also grants researchers access to supercomputing computing resources, robotics and neuromorphic platforms.
EBRAINS was built upon the foundation of Fenix, a research infrastructure established by five of Europe’s leading supercomputing centers: BSC in Spain, CEA in France, CINECA in Italy, CSCS in Switzerland and JSC in Germany. This infrastructure provides powerful computing, cloud services and data management to researchers.
Image from the Human Brain Project
The HBP’s work has significantly advanced personalized brain medicine and spurred the development of new treatments for neurological disorders. Key projects supported by EBRAINS include “The Virtual Brain” (TVB) for personalized brain simulations and MILEDI, a project focusing on understanding Alzheimer’s disease (AD) and identifying potential targets for new treatments.
It has also pushed forward technologies that mimic brain functions, enhancing fields like artificial intelligence (AI) and neuromorphic computing—a method that mimics the structure and function of the human brain to solve problems, recognize patterns and make decisions more efficiently.
With a substantial budget of EUR607 million, the project brought together over 500 scientists from 155 institutions across 19 countries. Using supercomputing infrastructures like Fenix, the HBP produced over 3,000 academic publications and developed more than 160 digital tools and medical and technological applications. Although the HBP officially wrapped up in 2023, the EBRAINS infrastructure remains a vital resource for the global scientific community.
Exascale Computing Project (ECP)
The Exascale Computing Project (ECP), an initiative by the U.S. Department of Energy (DOE), ran from 2016 to 2024 to create an exascale computing ecosystem. This initiative was designed to propel scientific discovery, ensure energy assurance, boost economic competitiveness and enhance national security.
Exascale computing systems are capable of performing at least 10^18 (one quintillion) floating-point operations per second (FLOPS. To put it in perspective, this is a thousand times more powerful than petascale supercomputers, which manage a quadrillion (10^15) calculations per second. Examples of petascale computers include Argonne National Laboratory’s Aurora (585 petaFLOPs) and Microsoft Azure’s Eagle (561 petaFLOPS).
One of the project’s major achievements was launching several state-of-the-art supercomputers, including Frontier, the world’s first exascale computer. Frontier debuted at Oak Ridge National Laboratory in May 2022 and was capable of executing 1.19 quintillion calculations per second—that’s a million trillion calculations—making it the world’s fastest supercomputer. Following closely is Aurora, which went online in June 2023, operating at 585 petaFLOPS, or 0.59 exaFLOPS.
Image from Singularityhub
The ECP has tackled critical challenges across various scientific fields. Notably, Frontier has supported NASA with several supercomputer simulation initiatives, including reducing jet aircraft noise, advancing green aviation and exploring deep-space star systems.
Recently, the DOE has partnered with the U.S. National Cancer Institute (NCI) to study cancer at cellular, molecular and population levels. Leveraging the ECP’s capabilities, NCI researchers have been able to analyze millions of cancer patient records. They conduct simulations at atomic, cellular and molecular levels to predict drug responses, optimize drug screening processes and enhance cancer drug discovery. This collaboration has significantly pushed forward the boundaries in the fight against cancer.
Blue Brain Project
The Blue Brain Project, launched in May 2005 at the École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland, is spearheaded by Professor Henry Markram and primarily funded by the Swiss government. Its ambitious goal is to reconstruct a digital model of the mammalian brain, eventually simulating the entire human brain..
The project uses powerful supercomputers like IBM’s BlueGene/Q, nicknamed Mira, along with BlueGene/L and BlueGene/P. These machines are essential for modeling the brain’s neural circuits across different scales—from individual molecules to entire brain regions. IBM’s Blue Gene system, which ended in 2015, aimed to build supercomputers capable of reaching petaFLOPS speeds while using minimal power. In 2018, it was upgraded to Blue Brain 5, a more advanced supercomputer developed by HPE, an edge-to-cloud platform-as-a-service company company.
The Blue Gene/Q supercomputer
Image from IBM
Set to conclude in 2024, the Blue Brain Project has already made significant contributions to our understanding of brain functions, disorders, and diseases. Some key achievements include:
- In 2015, they recreated and simulated a microcircuit consisting of neurons and synapses from the neocortical column—the neocortex’s “CPU” and launched a collaboration portal for this microcircuit.
- That same year, they developed an algorithm that located nearly 40 million connection points between neurons in a small sample of brain tissue.
- In 2016, a team led by Eilif Muller won significant supercomputing time from the U.S. Department of Energy’s Incite Leadership Computing Program. This award supports their work on simulating synaptic plasticity, crucial for understanding how brain activity affects synaptic connections
- In 2019, they tackled a long-standing mathematical challenge by accurately modeling how neurons grow, aiding research into genetic changes in the brain.
As the project nears its end, it plans to launch the Blue Brain Open Platform to offer the neuroscience community access to an extensive array of data, software, algorithms and biologically detailed brain models that have been developed since the project’s inception.
Throughout its duration, the Blue Brain Project has made remarkable progress in tackling the complex challenges of biological modeling, informatics and computational neuroscience. It has developed vital algorithms and software solutions necessary for simulating intricate brain structures, including neuron morphologies and synaptic connections, using supercomputing capabilities.
CERN’s Large Hadron Collider
3D cut of the LHC dipole
Image by CERN
The Large Hadron Collider (LHC), launched in 2008 at the European Organization for Nuclear Research (CERN), is one of the most significant particle physics projects ever undertaken. Located underground near the Franco-Swiss border, the LHC features a 27-kilometer ring of superconducting magnets. These magnets accelerate two beams of particles in opposite directions to nearly the speed of light before colliding them. The LHC relies on supercomputers to manage and analyze vast amounts of data from these particle collisions, leading to revolutionary discoveries.
One of its most historic achievements came in 2012 when the LHC confirmed the existence of the Higgs boson. Analyzing petabytes of collision data to reach this discovery was only possible thanks to supercomputing power, particularly from the Worldwide LHC Computing Grid (WLCG)—the largest distributed computer grid in the world, which spans 40 countries and 174 facilities.
Currently, the LHC operates several detectors to capture the collisions and interactions of billions of particles. One such detector, ATLAS, processes its data through powerful supercomputers, including Aurora and Vega, a petascale EuroHPC JU machine hosted at the Institute of Information Science in Maribor, Slovenia. These tools are crucial for sifting through the colossal datasets generated by experiments, helping to unlock further mysteries of the universe.
Image from Intel
Square Kilometre Array (SKA)
First envisioned in 1991, the Square Kilometre Array (SKA) is an international effort to build the world’s largest radio telescope. With a total collecting area exceeding one square kilometer and a budget of EUR 2 billion, the project is headquartered in the UK. Initial construction began in 2022, with the telescope expected to become operational by 2028.
Once completed, the SKA will allow astronomers to observe the sky with unparalleled detail and survey the entire sky much more quickly than any existing system. This advanced capability will help scientists explore how our galaxy formed, examine the Milky Way in depth and enhance our understanding of various cosmic phenomena.
Led by the University of Cambridge, the project brought together nearly 40 institutions in 11 countries to build two supercomputers—one in Cape Town, South Africa and another in Perth, Australia. These supercomputers will use specialized machine-learning algorithms to quickly tell the difference between important data and background noise. By 2019, the team had finished the design work for one of these supercomputers.
The future of supercomputers
From brain research and particle physics to astronomy and exascale computing, supercomputers are enabling breakthroughs that once seemed unimaginable.
In 2023, the global high-performance computing market was valued at a hefty US$50.02 billion, according to Fortune Business Insights. And it’s only getting bigger. Expected to grow at a rate of 9.2% annually, it’s projected to soar to US$109.99 billion by 2032. This rapid growth highlights just how crucial supercomputing is becoming in solving complex problems and advancing our understanding of the world.
Looking ahead, the role of supercomputers is only set to expand. These powerful machines will continue to help us explore new frontiers and solve some of the most critical projects that will shape our future.
Also read:
- Meet the World’s Top 10 Most Powerful Supercomputers
- 5 Uses of Supercomputers for Startups and Entrepreneurs
- How Supercomputers Are Affecting the Environment
- Tech Giants Racing to Unleash the Power of Quantum Supercomputers
Header Image by IBM
