ESiWACE

ESiWACE

Preparing for Exascale

The main goal of ESiWACE is to substantially improve efficiency and productivity of numerical weather and climate simulation by supporting the end-to-end workflow of global Earth system modelling in current and future high performance computing environments - especially those which will require the next generation of pre-exascale and then exascale computers.

Our primary interest in ESIWACE is around issues of data handling: in the initial compute systems where we hit the limitations of bandwidth to the POSIX file system; in post-processing compute systems where volume provides issues; and in the distributed systems (including cloud) which are necessary for multiple modelling groups to share data efficiently at the European and Global scale.

Three types of computing
The three domains of interest for data handling in ESIWACE.

People

Bryan Lawrence
Professor of Weather and Climate Computing
Julian Kunkel
Lecturer in Computer Science