Sweco Finland uses the LUMI supercomputer for demanding built-environment simulations

Sweco, Europe’s leading engineering and architecture consultancy, aims to build more sustainable societies together with its clients. Sweco Finland uses high-performance computing (HPC) and the LUMI supercomputer to enhance its special expertise in demanding computing. This brings time and cost savings in client projects.

Eero Kokkonen, Department Manager and Principal CFD Engineer works in Sweco Finland’s Special Services Business Unit that provides computational fluid dynamics (CFD) expertise to help clients achieve their goals.

Computational fluid dynamics can be used to simulate the flow phenomena of any gas or liquid. By simulating flow rates, pressures, and other turbulence quantities, it is possible to improve the energy efficiency of data centers, the safety of hydrogen power plants, indoor environment conditions in residential and office buildings, wind comfort in the built environment, or the effectiveness and environmental friendliness of fish passes. With CFD, a virtual prototype corresponding to physical reality can be produced in the early stages of design. This reduces risks in decision-making and guides the design in the right direction.

Computational fluid dynamics and, in particular, the use of high-performance computing, requires expertise that companies typically do not have. Supercomputers have mainly been tools for the academic community and research organizations. Since 2022, European supercomputer LUMI has enabled the broader use of supercomputers to support Finnish companies’ businesses.

LUMI is one of the world’s most powerful supercomputers located in CSC’s Kajaani data center. Kokkonen got to know CSC in his student days as he used CSC’s national supercomputers as part of his Master’s thesis work, and he has kept up this connection in the form of various client projects. Today, the department headed by Kokkonen uses the LUMI supercomputer to enhance the efficiency of computational fluid dynamics.

Clients benefit from modern technologies

Sweco has carefully evaluated and confirmed that the LUMI supercomputer is a secure, cost-effective, and flexible high-performance computing (HPC) environment that meets Sweco’s clients’ requirements.

– LUMI plays a significant part in our toolkit as we wish to offer the benefits of modern technologies and the best possible tools for solving our clients’ challenges that require very specific expertise. In the spirit of sustainable development, our principle is getting it right the first time, Kokkonen describes Sweco’s goals.

Getting the results of computational fluid dynamics typically takes several days or even weeks. In practice, the LUMI supercomputer enables up to ten times higher efficiency and the use of more accurate models.

– It is indeed hard to imagine CFD without a high-performance computing environment, says Kokkonen, an experienced user of HPC technology.

Cost-effective computing capacity and expert support

Kokkonen explains that there are differences between cloud services and high-performance computing when it comes to the hardware used, for example.

– Cloud services are intended for much smaller computing tasks, where each area of computing efficiency need not be optimal. Solutions specifically optimized for HPC minimize the bottlenecks of computing. This way, the resources required for CFD – in practice, several hundred processor cores – can be used for a large number of simultaneous computation tasks, Kokkonen explains.

– For our clients, this inevitably means significant added value in terms of design quality, schedules, and costs, Kokkonen continues.

While Kokkonen does not use cloud services in his work, comparing the features and costs of different services and products is necessary when looking for the best tools. This is why Kokkonen recommends that one should compare their options and purchase the best price-quality ratio to meet specific needs, also when it concerns technical computing. According to Kokkonen, the price of LUMI’s computing capacity is highly competitive.