Unlocking the universe: simulating cosmic strings with LUMI

Cosmic strings are hypothetical objects that may have formed in the early universe. Although no direct evidence of them has been found, scientists believe that these strings could reveal crucial insights into the nature of space-time and the forces that govern the cosmos.

Studying cosmic strings is challenging because they form complex, non-linear systems that can’t be solved with simple equations. This is where high-performance computing (HPC) steps in. Using the immense power of the LUMI supercomputer, researchers José Ricardo Correia at the Institute of Physics at the University of Helsinki, Finland and Carlos Martins at the Centre of Astrophysics at the University of Porto, Portugal, are running detailed simulations to better understand how cosmic strings evolve and what gravitational waves they might produce. With these detailed simulations, researchers hope to predict what signals future experiments, like the European Space Agency’s (ESA) gravitational wave detector LISA (Laser Interferometer Space Antenna), should look for. These signals could reveal entirely new physics, similar to how gravitational wave discoveries expanded the understanding of the universe.

Correia is showing me a simulation of cosmic strings done with LUMI’s resources. Pieces of string are moving on the researcher’s screen: changing forms, creating different shapes, sometimes creating a loop, sometimes disappearing. He is working closely with his colleague Martins on this project named GRaCoS: Gravitational Radiation from Cosmic Strings.

– It’s very much possible that in the early universe, as it cooled, wire-like objects likely formed. The process is similar to when water cools down over time and eventually freezes into ice. This doesn’t happen completely homogenously – ice tends to have some cracks. If these cracks are essentially like wires, we call them cosmic strings. These strings are actually expected to emit gravitational waves in different processes. Our research focuses on how cosmic strings evolve over time and the resulting effects, such as the amount of gravitational waves they emit, Correia explains.

Understanding the universe

But why do cosmic strings matter? Although they are currently purely theoretical, their potential existence is intriguing because they could have significant implications for our understanding of the universe.

– We have a sort of “cookbook” of what are the constitutions of the universe, so we know the particles in essence, but there are a few things that probably are missing and need to be added. We know that there should probably be something beyond the current theories, but we don’t know what exactly. If the existence of cosmic strings were proven, it would reveal conditions beyond the standard model of particle physics and standard cosmology.

This information would, for example, advance our knowledge about the evolution of the early universe and much more.

Image: Researcher José Ricardo Correia is working at Institute of Physics at University of Helsinki.

HPC to the rescue

Nobody has observed cosmic strings yet. But different kinds of observational data from space does exists, e.g., from ESA’s former Plank satellite. The detection of cosmic strings is a primary target for ESA’s future space-based missions, such as LISA, which are set to launch about ten years from now.

The way to advance our knowledge about cosmic strings is to make simulations with HPC systems and compare these simulations to physical models. Correia’s colleague Martins at the University of Porto explains:

– Studying cosmic strings is difficult because they interact in complex ways. If there was just one string, it could be solved with simple equations, but when many strings interact, it becomes too complicated. This is why we need HPC.

– Simulations help calibrate models that predict gravitational wave signals and other observational outcomes. This project runs higher-resolution simulations than before, thanks to LUMI’s computing power. With better simulations, we can create more accurate models and make better predictions to compare with future observational data, Martins explains.

The team’s work improves our understanding of the potential signatures that cosmic strings could leave in observational data. If someone detects a signal that can’t be explained by standard physics, the question becomes: what caused it?

– To answer that, you need to know what potential new physics might look like, similar to how scientists predict what dark matter signals could look like, even though we don’t know what dark matter is. Our work helps create that “portrait” for cosmic strings, so if future experiments like LISA detect a certain signal, it could point to the existence or properties of cosmic strings. We are exploring possibilities and gaining a better sense of what to look for, Correia explains.

LUMI – one of a kind

The research group had used GPU-based computing earlier but first they needed to get things running with the new environment.

– The experience of using LUMI can be divided into two phases, so the first one is the part of starting to use LUMI. We needed to port the code to AMD GPUs, get familiar with toolkits, libraries and so on. For me, this was quite fun. After that it became “work as usual” – if you can call it so with one of the most powerful supercomputers in the world. We are trying to get as much statistical data as you can at a very high resolution in a very reasonable amount of time. We are stretching LUMI to its limits, but none of those runs take a very large amount of time, so a singular run takes probably less than an hour. For our work, LUMI’s computing power is crucial – this would likely be impossible on any other European HPC system. LUMI is exceptional, it’s one of a kind, Correia concludes and thanks also the LUMI User Support Team for their support.

Have a look at the video interview below where Correia tells more about their research:

Author: Anni Jakobsson, CSC

Image on top: Screencap taken from a cosmic string simulation. Copyright: José Ricardo Correia and Carlos Martins