Cornerstone Technologies
Cornerstone Technologies contain efforts to improve and expand our e-Science toolbox to solve problems faster and more efficiently, regardless of the area of the application.
Current projects:
Big data processing for experimental collider physics
Long-term project, period 2010–present
PI/contact: Rebeca Gonzalez Suarez
University: Uppsala University
Experimental Collider Physics
%20(2)%20(2)%20LHC-and-space-bigdata.jpg)
Applied scientific computing and High-performance computing
In Computational Science, we develop methods and algorithms to model, simulate, and analyze complex systems with high precision and performance. The research ranges from fundamental numerical algorithms and uncertainty quantification to tailored solutions for fields such as quantum physics, life sciences, and fluid dynamics -- often in combination with advanced computing environments such as cloud platforms and high-performance computing. The focus is on reliable, efficient, and scalable methods for data-driven and computation-intensive applications
Long-term project, period 2010–present
PI/contact: Stefan Engblom
University: Uppsala University
Applied scientific computing
High-performance computing
Simulated noise map of the top floor at Ångströmlaboratoriet. A low-frequency acoustic point source is placed in one of the offices, walls in black.
Data science and intelligent decision making
Data Science addresses the acquisition, storage, and analysis of observed complex data, towards detecting patterns and making predictions. Intelligent Decision Making addresses the modelling of constrained decision-making problems, towards running a model for input data on an off-the-shelf solver. In this programme, we develop e-science methods to support the whole pipeline: from data to decisions.
Long-term project, period 2010–present
PI/contact: Pierre Flener
University: Uppsala University
Data Science: Creating value through data
Optimisation: The Science of Taking Better Decisions
Zoom imageA map of online interactions between social media accounts, used to identify patterns such as political polarisation.
AI2TWIN – retrieving digital twins using physics-informed AI for in-situ and operando imaging reconstructions
This project aims to streamline the development of digital twins from complex in-situ and operando experiments carried out at Large-Scale Facilities such as MAX IV in Lund.This involves the development of: i) new AI-based Physics-Informed Neural Networks (PINNs) to reconstruct the underlying physical laws directly from experimental datas and ii) optimal structure-preserving Finite Element Methods (FEMs) to test and transfer those laws into actual digital twins.
Period: 2025–2026
PI/contact: Robert Klöfkorn
University: Lund University
Flow problems in porous media: Modelling, approximation and implementation
In this project, we want to combine these three major objectives:
(A) Development of a model for the assessment of the performance of groundwater remediation. (B) Construction and analysis of modern time-stepping methods. (C) Efficient implementation of the time-stepping methods for relevant problems.
Period: 2023–2025
PI/contact: Monika Eisenmann
University: Lund University
High-Performance and Automatic Computing*
Long-term project, period 2010–present
PI/contact: Paolo Bientinesi
University: Umeå University
High-Performance and Automatic Computing
The goal is to provide domain experts with tools for the generation of
high-performance algorithms and code with minimal or no human intervention. The focus is on domain specific languages, compilers, and libraries for matrix and tensor computations.
*Formerly "Parallel and scientific
computing: theory, algorithms and software
Virtual IT Infrastructure (clouds & grids) for eScience
Long-term project, period 2010–present
PI/contact: Erik Elmroth
University: Umeå University
Cloudresearch
The (semi-)autonomous management infrastructure and applications for distributed computing infrastructures is in scope for this project. Challenged by the ever increasing scale, complexity, and need for efficiency and reliability, the means are to increase the level of abstraction for infrastructure owners, application providers, and end-users while empowering eScience research with the the next generation methods and tools.
Multiscale Modeling of Dynamical Systems
Period: 2023–2024
PI/contact: Mats G. Larson, Karl Larsson
University: Umeå University
Computational design optimization of devices for wave propagation in lossy materials
Period: 2023–2024
PI/contact: Martin Berggren, Emaldeen Hassan
University: Umeå University
A novel method to transform the modeling of global lake
ecosystem dynamics using Model Order Reduction
Period: 2025–2026
PI/contact: Cristian Gudasz (UMU), Philipp Birken (LU), Mengwu Guo (LU)
University: Umeå University
Accurate volumetric reconstructions from
state-of-the-art observations
Period: 2025–2026
PI/contact: Maria Hamrin (UMU), Juan Carlos Araujo-Cabarcas (UMU), Eddie Wadbro (Karlstad), Jakub Vaverka (UMU)
University: Umeå University