Psyclone Development and Support 2026-27

Location

Description

PSyclone is a specialist scientific software tool developed by the UK's Science and Technology Facilities Council (STFC) for high-performance computing (HPC). It's a code-generation and optimisation framework that helps scientists run complex simulation models efficiently on modern supercomputers (CPUs, GPUs, clusters, etc.) Over the past decade, PSyclone has been a valuable software asset, enabling collaborations for the UK Met Office to enhance their use of HPC and extreme-scale computing. This is essential for our growing requirements to execute larger and more accurate weather and climate simulations. In this project we will continue to develop features of PSyclone to support the LFRic, NEMO and Data Assimilation components of the Met Office's Momentum software stack. This work will include support for new science-led features as well as performance optimisations for multi-core CPUs. It will also continue to look at performance optimisations specific to the CPUs with which the Met Office's new Phase1b supercomputer is equipped.

Award Detail

CPV Codes

• None found

Legal Justification

The supplier Science and Technology Facilities Council (STFC) is government-owned. The Met Office have collaborated with STFC for 10 years on the development of the software tool PSyclone. PSyclone is a domain specific language that enables performance portability of scientific codes on next generation compute architectures like GPUs. PSyclone is a key underpinning component of LFRic, our next generation atmosphere model which is central to our Next Generation Modelling Systems Strategic Action. Furthermore, these capabilities are also important in the preparations for Generation 2 supercomputer procurement. PSyclone is a specialist scientific software tool developed by the UK's Science and Technology Facilities Council (STFC) for high-performance computing (HPC). It's a code-generation and optimisation framework that helps scientists run complex simulation models efficiently on modern supercomputers (CPUs, GPUs, clusters, etc.). The project is cutting-edge research, and ongoing work is strongly dependent on background knowledge of the key workers at STFC. The collaboration is successful and productive.

Other Information

** PREVIEW NOTICE, please check Find a Tender for full details. **