Finnish Steerprop leverages the LUMI supercomputer for advanced marine technology

When ships are needed for extreme conditions requiring reliable, energy‑efficient, and powerful propulsion systems, attention turns to Finland. The Finnish propulsion cluster is known for world‑leading manufacturers and high‑quality RDI. This requires the best tools, and that is why Steerprop uses the LUMI supercomputer.

Steerprop is one of the companies in the Finnish propulsion cluster, with strong expertise in designing and manufacturing advanced marine propulsion systems for demanding conditions. The company is among the global leaders in its field and operates through a distribution network in over 50 countries. Founded in 2000 and based in Rauma, Finland, the company specialises in tailor-made solutions such as azimuth thrusters, used in vessels including icebreakers, ferries, and offshore ships.

In the maritime industry, azimuth refers to the capability to generate omni-directional thrust by rotating a thruster unit around the vertical axis. The thruster provides both propulsion and steering for the vessel, effectively replacing the traditional shaft line with propeller and rudder arrangement.

Mahish Mohan serves as Chief Hydrodynamicist at Steerprop. His responsibilities include propeller design and computational fluid dynamics (CFD) simulations. He joined Steerprop in 2018, influenced by Finland’s strong concentration of propulsion industry expertise.

Using his CFD expertise, he develops the company’s capabilities in applying computational tools for tailored customer solutions. In addition, he is pursuing a part-time PhD at Aalto University.

Discovering LUMI through a Business Finland project

Mohan became familiar with the LUMI supercomputer when Steerprop participated in Business Finland’s Necoprop research project. During the project, the team explored high-performance computing options, and LUMI supercomputer quickly emerged as a strong candidate.

Collaboration with LUMI began with a Try&Buy project, which allowed Mohan to evaluate LUMI’s suitability for his simulation needs.

– The free of charge Try & Buy project includes CPU, GPU, and storage resources for testing, as well as expert support for implementation, says Morthen Mathisen, CSC’s Customer manager. – After a 3–6 month trial period, the company can either continue using the system under a service agreement or stop after the test phase.

The experience was very positive for Steerprop.

– I’m very happy with LUMI. It fits our use cases perfectly, Mohan says, highlighting one key technical advantage.

– LUMI’s compute nodes have a large number of cores. This significantly benefits parallel computing, especially with the OpenFOAM software used by Steerprop. Fewer nodes mean less inter-node communication, which speeds up simulations.

The onboarding process went smoothly thanks to Mohan’s prior HPC experience. Although some system-specific differences had to be learned initially, LUMI’s support services proved efficient. Software updates and technical support were delivered quickly, often within a few days. Assistance was also provided for compiling custom-developed software for LUMI.

Following the pilot phase, Steerprop moved into commercial use of LUMI.

Faster computing means faster delivery and higher quality for customers

Although Steerprop has its own in-house computing cluster, its capacity is limited. Large simulations or multiple concurrent runs quickly consume available resources. This leads to compromises: simulations must either be slowed down or mesh resolution reduced, which can negatively affect accuracy and business outcomes.

LUMI addresses this challenge by enabling rapid scaling of computing resources, large-scale simulations, and multiple parallel runs. As a result, engineers can use more accurate models without sacrificing performance.

The benefits of LUMI are most visible in customer projects. The propeller design projects typically operates under tight schedules, often around 3–4 weeks.

– Individual simulations can take several days, says Mohan. – With LUMI, simulation times are significantly reduced. For example, a job that would take 5–6 days on our own cluster can be completed in under two days on LUMI.

This enables testing more design variations, achieving better optimization, and maintaining project schedules. Ultimately, customers benefit from higher-quality solutions delivered on time.

Mission: leading position and zero-emission maritime industry

Steerprop, together with other Finnish operators, also aims to achieve zero emissions in shipping and to maintain Finland’s leading position as an exporter of clean marine technology and Arctic shipbuilding expertise.

– State-of-the-art CFD capability is in the centre to meet the market expectations set for Steerprop. Using LUMI we ensure that we have high performance computational resources available even for the most demanding simulation cases, says Tero Tamminen, Technology Manager at Steerprop.

As one of the fastest, but also the greenest supercomputers in the world, LUMI is a perfect fit for Steerprop’s and clean marine technology needs. LUMI’s and its data center’s minimal environmental impact, cost-effective energy consumption and the utilization of excess heat from supercomputer cooling in Kajaani district heating, make LUMI a supercomputer that supports the green transition.

Growing computational needs bring challenges

Steerprop plans to expand its use of LUMI in the future, particularly for increasingly complex and computationally demanding projects. One key development area is acoustic modeling, driven by growing requirements related to underwater noise and environmental impact.

– Acoustic simulations are significantly more demanding than traditional CFD, Mohan notes. – They require additional solver stages and heavy post-processing.

Mohan estimates that future simulations may require up to 1,000 cores or more. LUMI provides a platform where such scaling is possible.

Although he is very satisfied with LUMI, there are still areas for improvement.

– Currently, only one user at Steerprop – myself – uses LUMI. We are actively evaluating whether LUMI usage should be expanded to support multiple simultaneous users and projects at Steerprop. I see this as a challenge, Mohan tells.

The main challenge relates to resource availability. After intensive workloads, users may experience delays when trying to access the queue again.

LUMI uses Slurm (Simple Linux Utility for Resource Management), a widely used workload and resource management system in supercomputers and clusters. It manages job scheduling, resource allocation (such as CPU, GPU and memory) and execution on LUMI, ensuring that multi-user workloads run efficiently and fairly across different computing units.

– We fully understand the fairness of the Slurm-based queuing system, but it can make it difficult to use LUMI when schedules are tight and predictability is critical, Mohan muses. – The challenge can be accentuated if multiple users from the same organization are simultaneously dependent on the system.

LUMI user support is aware of the challenge. It is currently being investigated whether the national Roihu supercomputer—launched in May 2026—and the even more powerful EuroHPC supercomputer LUMI-AI—scheduled for launch in late 2027—could offer a solution also for industry computing capacity.

Image on top: Adobe Stock