Computer Scientist / Numerical Analyst for research into exascale
- Vacancy No.: VN16-40
- Department: Forecast Department
- Grade: A2
- Section: Development Section
- Job Ref. No.: STF-PS/16-40
- Reports to: Team Leader
- Publication Date: 21 November 2016
- Closing Date: 18 December 2016
ECMWF is both a research institute and a 24/7 operational service, producing and disseminating numerical weather predictions to its Member States. ECMWF carries out scientific and technical research directed to the improvement of its forecasts, collects and processes large amounts of observations, and manages a long-term archive of meteorological data. Satellite and in situ observations provide the information for up-to-date global analyses and climate reanalyses of the atmosphere, ocean and land surface. For details, see www.ecmwf.int/. In 2014, ECMWF initiated a ten-year research programme on scalability in high-performance computing (HPC). The programme is funded partly by ECMWF’s Member States and aims to significantly improve the time- and energy-to-solution of numerical weather prediction systems. ECMWF uses its HPC facility to produce a time-critical, twice-daily global numerical weather forecast. This operational forecast generates very large amounts of input/output (I/O) over very short bursts, reaching tens of terabytes (TB)/day, from which millions of user-defined products are generated. Considering the likely growth of the overall forecast system, it is estimated that by 2020 the operational model will output 128 TB/day and need to archive 1.25 petabytes (PB)/day. The current I/O workload is one of the strongest bottlenecks in ECMWF’s workflow and I/O is recognised as one of the main challenges on the route to exascale computing. As part of the EU H2020 funded NEXTGenIO project, ECMWF is researching how best to exploit emerging memory extension technologies for its workflow, in particular to generate weather forecast products in a timely manner.
Summary of the role
ECMWF is recruiting a Computer Scientist or a Numerical Analyst to work on the NEXTGenIO project. The role involves integrating the I/O server layer in ECMWF’s Integrated Forecasting System (IFS) with the newly developed product generation software, leveraging dedicated storage device technology developed in NEXTGenIO . The aim is to minimize I/O in the time-critical IFS applications, to reduce duplication of compute tasks and to accelerate the time-to-delivery of weather forecast products.
Main duties and key responsibilities
- To research scalable data handling strategies for handling meteorological product generation involving significant I/O workloads To integrate the IFS I/O server with the newly developed product generation software
- To employ NVRAM technology to maintain model output data in memory for successive product generation
- To implement mechanisms to ensure the overall system’s resilience to failure
- Excellent analytical and problem-solving skills with a proactive approach, together with an interest in identifying, investigating and solving technical problems
- Dedication and enthusiasm to work in a small team
- Excellent interpersonal and communication skills, listening to and respecting the views of others
- Ability to work under pressure
Qualifications and experience required
- A university degree, or equivalent, in a discipline related to computer science, physics, mathematics or engineering is required. A PhD is desirable but not essential.
- At least three years of experience developing parallel codes on highperformance computing facilities, including distributed systems;
- experience with interfaces from Fortran to C/C++ would be an advantage.
- Experience with developing and maintaining large scientific codes would be an advantage, in particular implementing numerical algorithms.
Knowledge and skills (including language)
- Very good knowledge of object-oriented programming and several years of experience developing with C/C++ and MPI.
- Strong understanding of I/O and communication on HPC systems. Good understanding of numerical methods.
- Strong background in UNIX systems (and/or Linux).
- Candidates must be able to work effectively in English and interviews will be conducted in English.
- A good knowledge of one of the Centre’s other working languages (French or German) would be an advantage.
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 £55,488 net of tax. This position is assigned to the employment category STF-PS 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: 15 months (or longer if funds available), but not beyond 30 September 2018
- Location: The role will be based in the Reading area, in Berkshire, United Kingdom.
How to apply
Please apply by completing the online application form available at www.ecmwf.int/en/about/jobs.
ECMWF has an Equal Opportunities Policy and applications from all suitably qualified candidates are welcome.
Staff are usually recruited from among nationals of the following Member States and Co-operating States: Austria, Belgium, Bulgaria, Croatia, Czech Republic, Denmark, Estonia, Finland, former Yugoslav Republic of Macedonia, France, Hungary, Germany, Greece, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Montenegro, the Netherlands, Norway, Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey and the United Kingdom. Staff from other countries may be considered in exceptional cases.
- Organization: ECMWF - European Centre for Medium-Range Weather Forecasts
- Location: Reading
- Grade: A-2, Grade
- Information Technology and Computer Science
- Physics and Mathematics
- Meteorology, Geology and Geography
- Closing Date: 2016-12-18