Our mission is to empower researchers and industrial partners to produce high quality scientific software and machine learning solutions that run optimally on state-of-the-art high performance computers. We collaborate with researchers at the University of Cambridge as well as other National and International institutions, pursuing the objective "Better Software for Better Research".
Services we offer
- Software engineering consulting
- We design, extend, and refactor scientific software in all subject areas
- Analyse and optimise existing scientific software
- Modernise legacy software
- Migrate existing applications to HPC
- Accelerate existing applications with GPUs
Recent exemplar projects
- Modernisation and GPU acceleration of the PROMPI stellar evolution code (via DiRAC)
- RADDISH: Implementation of a crowding agent based model in C++, OpenMP and MPI
- CVODE support in FEniCS
- Optimal parallelisation in CASTEP
- Optimisation CASTEP on Intel's Knight's Landing (KNL) Platform
- ExoAI: exoplanet discovery via deep learning
- Distributed-memory parallelisation of TROVE (Theoretical ROVibrational Energies)
Recent reports and publications
- Quantification and Selection of Ictogenic Zones in Epilepsy Surgery
- Pushing the Limits of Exoplanet Discovery via Direct Imaging with Deep Learning
- A scalable and modular Material Point Method for large-scale simulations
- SODECL: An Open Source Library for Calculating Multiple Orbits of a System of Stochastic Differential Equations in Parallel
- OpenFOAM on Intel Xeon Phi Processors