LUMI SmartSim Workshop

You are welcome to join us in 1-day hands-on workshop “LUMI SmartSim”. Using machine learning (ML) within HPC applications offers the advantage that it can improve the overall efficiency of HPC workflows by allowing ML predictions to help guide or replace parts of simulations.  However, there are challenges in coupling the two approaches as HPC applications are typically written in C, C++ or Fortran and run on CPUs, whereas ML frameworks are usually Python based and benefit from being run on GPUs. In this tutorial we will introduce SmartSim, an open-source framework that has been developed to help address these challenges and allow researchers to easily adapt their traditional HPC workloads to include ML. This is done with a distributed orchestrator, which allows data to be exchanged between the HPC application and the ML model.

Event Info

Dates: 31 March 2026
Time: 10:30 – 16:00 CEST
Location: Room 2.03, SURF Utrecht, Kantoren Hoog Overborch (Hoog Catharijne), Moreelsepark 48, 3511 EP Utrecht
Organizer: HPE, LUMI User Support Team (LUST), SURF Utrecht

Target group

LUMI users interested in integrating machine learning into existing HPC workflows.

Learning outcomes

In the morning session, we will consider how ML and HPC applications may be coupled together and then give an overview of SmartSim and show how it can interact with simulation data in HPC applications in order to integrate ML models. We will show real examples of how SmartSim has been used in different fields of research such as Computational Fluid Dynamics and Climate Modelling. We also explain how SmartSim can be integrated within an application and how to deploy the resulting workflow.

The afternoon session will take a more hands-on approach where attendees can run examples of hybrid scientific simulation/AI workflows using SmartSim.  Attendees will learn how to launch SmartSim leveraging the system’s native batch scheduler and job launcher and gain a practical understanding of the modifications required to adapt HPC applications to use ML. We will also provide examples where attendees can run real applications that utilise SmartSim.

Agenda

10:30 Overview of ML and HPC coupling and introduction to SmartSim
11:30 Practical use of SmartSim
12:00 Lunch
13:00 Connecting to LUMI
13:30 Hands on practical examples
15:00 Open session: Discuss using SmartSim in user’s own applications or run more examples
16:00 Finish

Registration

Deadline: 23 March 2026, at 23:59 CEST

Click here to register for the workshop. 

Please register by the deadline to secure your place. Spaces are limited and available on a first-come, first-served basis. Participation in the workshop is free of charge and includes coffee/tea/water and a few snacks during the days. Please be prepared to cover your own lunch costs.

We have 48 in-person seats and every participant will receive a quick confirmation on a first-come, first-served basis so that they can arrange travel.

Attendees are encouraged to bring their own applications to try with SmartSim in the afternoon session or be prepared consider how SmartSim might be used for their own workflows/applications.

Smartsim is portable to many HPC systems and has been run on laptops and large clusters. The machine learning backend supports both Nvidia and AMD GPUs and CPU backends are provided for Apple (both Intel and Apple Silicon) and Linux (x86_64). On HPC systems the SmartSim infrastructure library can integrate with Slurm, PBS and LSF.