ECMWF – Scientist – Scalability Programme
ECMWF, Shinfield Park, Reading, UK
Closing date: 24 January 2019
Summary of the role
In 2014, ECMWF embarked on a ten-year research programme on HPC Scalability. This programme is funded partly by ECMWF’s Member States and aims to achieve extreme-scale numerical weather prediction capability by 2025. As part of the Scalability Programme, ECMWF is engaged in partnerships with its Member State organisations, academia, national HPC centres and hardware vendors to prepare for the extreme-scale challenge, in terms of both computing and data handling. The programme explores advanced developments of numerical methods, algorithms, programming models and workflows to substantially enhance efficiency and scalability of the entire forecasting system on future HPC architectures.
These developments are further supported through ECMWF’s partnership in several projects that are externally funded by the European Commission through its Horizon-2020 programme, for example ESCAPE-2, ESiWACE(-2), EPiGRAM-HS, EuroEXA.
ECMWF is now looking to recruit a scientist who will work on the implementation of novel software developments throughout the Integrated Forecasting System (IFS), aiming to benefit from recent advances on scalable algorithms and data structures by integrating these into the IFS. The work will be located at the interface between computational and applied science, closely working with scientists in the Earth System Model and Earth System Assimilation Sections, and will thus cover computational aspects of both the forecast model and the data assimilation system.
The position sits within the Integrated Forecasting Systems Section at ECMWF and will report to the Code Efficiency Team Leader.
Main duties and key responsibilities
S/he will perform scientific and technical research and development in the following areas:
- Implementation of a generalised approach to representing model fields in assimilation and forecasts, their mapping to observation locations, performing domain-distributed and load-balanced operations on such fields (based on ECMWF’s Atlas library).
- Implementation of the coupling and concurrent execution of Earth-system components such as the ocean and sea-ice to enhance parallelism, also exploiting multiple grid capabilities.
- Continuing the implementation and performance optimization of reduced precision arithmetics.
The successful candidate will be recruited at the A2 grade, according to the scales of the Co-ordinated Organisations and the annual basic salary will be £59,228.40 NET of tax. This position is assigned to the employment category STF-PL as defined in the Staff Regulations.
Full details of salary scales and allowances are available on the ECMWF website at www.ecmwf.int/en/about/jobs, including the Centre’s Staff Regulations regarding the terms and conditions of employment.
Starting date: As soon as possible.
Length of contract: Four years, with the possibility of renewal.