This is how a quantum computer is built

The chapel of the Barcelona Supercomputing Center has a new tenant. This deconsecrated former church, which has housed the MareNostrum 1, 2, 3, and 4 supercomputers since 2005, is now taking an exponential leap forward to become the home of the first publicly accessible Spanish quantum computer built with 100% European technology.

The Quantum Spain project, tasked with bringing this initiative to life, has a budget of €22 million to support the construction of the computer, the creation of the ecosystem to make it accessible, and research into quantum algorithms, a crucial component in advancing this innovative technology.

Building on the mission of the MareNostrum projects, the goal is to provide Spanish research with a tool that promises to revolutionize our understanding of the world in the not-too-distant future, opening up new possibilities in fields ranging from drug design and materials physics to finance and artificial intelligence.

But what exactly can a quantum computer contribute, and how does it differ from a traditional one? “It’s not easy to explain in a single sentence,” says Alba Cervera, the physicist in charge of coordinating the project. “The quick answer is that a quantum computer is a computer that works according to the rules of quantum mechanics, but then we would have to explain what those rules are, and that’s when it gets complicated.”

To understand the difference between a quantum computer and a traditional one, you have to understand the difference between bits and qubits. A bit is the basic unit of information used to write the language that governs conventional computers, and in binary code, it can be expressed as either a 0 or a 1.

A qubit has other types of properties tied to do with the quantum world, in which one thing can be several things at once. Thus, instead of a 1 or a 0, a qubit can be a 0, a 1, or a mixture of both: what is called a superposition of states. But in addition to this, qubits can communicate with each other, so that the signal sent to one qubit affects the others, utilizing another quantum property called entanglement.

“This makes their operation completely different from that of traditional bits and means that the algorithms used to program them must also be of a completely different nature,” explains Cervera.

It’s not, therefore, a question of computing power, but rather the type of operations quantum computers can perform. Comparing the computing power of the MareNostrum 4 supercomputer with that of the MareNostrum 5 is very simple: 13.9 trillion operations per second versus 314 trillion operations per second. When we enter the realm of quantum computing, the issue becomes more complex because these are radically different operations.

It’s true that qubits, with their ability to express a 0, a 1, and all the values in between, exponentially multiply the capacity of traditional chips. But it’s not just that. Entanglement — the way qubits influence one another — is important, but above all, it’s the fact that the algorithms that govern these calculations are of a fundamentally different nature. By utilizing quantum properties, these algorithms can eliminate many steps, allowing quantum computers to solve tasks in minutes that would take a traditional computer years.

“The most immediate example is cryptography,” explains Cervera, “that is, the encryption of information. One of the most widely used forms of cryptography is based on the product of two prime numbers. Multiplying two prime numbers is very simple. But if we are given the product of that multiplication, and we have to figure out which two prime numbers multiplied together give that result, the operation can be truly complex: a task that, depending on the size of the given number, can take thousands of years for an ordinary supercomputer to solve.”

In 1994, U.S. mathematician Peter Shor developed an algorithm that, making use of quantum properties, could solve this problem much more efficiently, turning those thousands of years into mere minutes. “It’s not a question of speed, although that’s the result, but rather a type of algorithm specifically designed to take advantage of quantum properties so that it can solve the problem using far fewer operations, resulting in a much shorter resolution time and also considerable energy savings,” explains Cervera.

The difficulty lies in the fact that quantum hardware is not yet capable of supporting this type of operation. In fact, optimistically, even for algorithms that have been shown to work theoretically — such as Shor’s algorithm — it will take at least 15 years before the technology can take advantage of them. The operation of a quantum computer is still very unstable, and many errors occur. The great challenge lies in correcting these errors to achieve reliable processes.

It’s been said that a quantum computer will be able to solve all of humanity’s problems because it’s capable of analyzing all solutions at the same time, but this isn’t the case either. It’s important to keep in mind that having all the answers is the same as having none. “What starts to matter from there is the ability to formulate the question in such a way that we obtain a useful answer,” says Cervera. “And that, in computer science, translates into designing the correct algorithm.”

The key lies in what is called “quantum supremacy,” in other words, what types of processes can be used to design an algorithm in which quantum technology is more efficient than the most powerful traditional computer. To start, this includes tasks that require simulating interactions in the quantum world: those involving particle physics.

“Developing a drug, for example, requires experimentation with molecules that are composed of atoms, which are composed of particles governed by the rules of quantum mechanics,” says Cervera. “A computer that runs on this technology will be much more suitable for carrying out these types of simulations than one that has to imitate it.”

The same is true for materials physics and the operations required for the development of artificial intelligence. It is believed that it could also be very useful in optimizing logistics processes such as shipping Amazon packages or financial problems such as optimizing investment portfolios, but these applications are still being studied.

To function optimally, a quantum chip needs to be kept in a vacuum to isolate it from any potential disturbances, and at a temperature close to absolute zero, or -273 degrees Celsius. The chip itself occupies no more than six square centimeters. All the equipment we see associated with a quantum computer is designed to create this vacuum and bring the temperature down to such extremely low levels.

Once operational, the quantum computer will become part of MareNostrum 5 and will join the Spanish supercomputing network, of which the Barcelona Supercomputing Center is both a member and coordinator. This will make it available for free to any researcher. Anyone needing to carry out experiments involving tasks that require this technology can submit a project proposal, which will be evaluated and approved by a committee of experts from the center.

We are, therefore, witnessing the most advanced technology in the world, while simultaneously being at the dawn of what this technology can offer. When we observe the brand-new five-qubit chip that has just been installed in the chapel of the Barcelona Supercomputing Center, we are actually witnessing a moment that will be studied in the future as a pivotal point in the history of quantum computing in Spain, much like the punch cards of the last century mark the origin of traditional computing. It is an initiatory moment, and only time will reveal where it may lead us.

Sign up for our weekly newsletter to get more English-language news coverage from EL PAÍS USA Edition