Blog Page


Europe's fastest supercomputer just connected to a quantum computer in Finland — here's why – Interesting Engineering

LUMİ Microsoft 
Last month, LUMI, the fastest supercomputer in Europe, was connected to HELMI, Finland's first quantum computer, a five-qubit system operational since 2021. This makes Finland the first country in Europe to have created such a hybrid system – it is one of the few countries worldwide to have done the same.
LUMI is famous – the supercomputer ranks third in the latest Top 500 list of the world's fastest supercomputer and can carry out 309 petaflops. LUMI, too became operational in 2021.
VTT Technical Research Centre of Finland worked with CSC and Aalto University, within the Finnish Quantum Computing Infrastructure framework, to make the connection between the computers, according to a release.
While quantum computers are extremely powerful at certain tasks, they need to be supervised by traditional classical computers and are challenging to use as standalone systems. According to, the ecosystem around quantum computing is not mature to offer a decent set of tools to program quantum computers, run programs, and read the output.
When such a system is created, it enables hybrid computing projects and "boosts the development of the required quantum algorithms and software". The potential of this technology to solve real-world use cases will soon arise.
"We aim to augment supercomputing with quantum computing," said Ville Kotovirta, leader of the quantum algorithms team at VTT, reported
"The most challenging problems for classical computing – for example, electronic structure problems – could hopefully be solved more efficiently with quantum computers in the future. On the other hand, supercomputers could help optimize quantum algorithms, and they can post-process measurement data to mitigate errors in the computation. As quantum computers get larger, these optimization problems get more challenging and could benefit from supercomputing," said Kotovirta.
In the future, quantum computers will solve a minimal set of functional problems that supercomputers would take ages, thousands of years, to solve. The former will be able to solve in minutes or hours. These problems include developing new products and materials, for example, in the pharmaceutical, chemical, and battery industries.
The connection relies on the secure HTTPS protocol between the endpoints, said Kotovirta. "Users develop a hybrid algorithm on the LUMI side and submit the job to the LUMI queuing system for execution. Quantum jobs are forwarded to HELMI securely over HTTPS, and the HELMI back end controls the hardware electronics to do the actual quantum computation," said Kotovirta.
After which, the result is returned to the LUMI side of the program that made the call. "The program might then combine the result with whatever classical computation it may have performed. Then it might start the next iteration, or it might display the results to the user," said Kotovirta.
The merged computing power can give rise to faster and more accurate machine learning applications to generate novel molecular structures, accelerating the process of new material design. 
Advantages can be expected in areas that demand high accuracy and quality of computational predictions, such as short-term weather prediction. It is imperative to predict thunderstorms, hurricane paths, and tsunami propagation accurately. 
"This is connected to real-time analysis of Earth observation data from satellites. In the long term, high-quality image processing could, for example, detect a budding forest fire before it spreads uncontrollably. In the financial sector, higher accuracy in algorithmic trading naturally comes with an instant reward," said the release.
Go inside the discovery of the largest species of bacteria ever found by modern researchers. T. magnifica will change textbooks.