Jeff Zais & Wolfgang Hayek, NIWA

“NIWA data movement”

Wednesday 15 February, 1.30pm – 2.05pm

Abstract

NIWA (the National Institute of Water & Atmospheric Research, Limited) is one of the New Zealand Crown Research Institutes, dedicated to the study of fresh water, oceans, weather and climate. This talk will introduce NIWA and highlight how various critical data is collected and used. This will include the path of data as it moves from collection points, through pre-processing steps, and into a variety of HPC simulations ranging from fish populations studies to climate simulations. Included will be some detail on the volume of data that is generated, planning to handle the data long term, and studies on potential future systems for handling the NIWA workloads.

Bio

Since 2019, Jeff Zais has been an HPC Architect at NIWA watching over the computing and storage systems used for weather forecasting, climate studies, and other research on the oceans and waters of New Zealand. Jeff started as an HPC user at Ford Aerospace, analysing the structures of weather and communication satellites for Ford Aerospace. This was followed by a series of positions at HPC vendors (Cray Research, IBM, Lenovo) with a focus on benchmarking, performance measurements and estimates, and system architecture. Jeff has degrees from the University of Wisconsin (B.S) and from Stanford University (M.S., Ph.D.).

Wolfgang Hayek is a HPC Research Software Engineer at NIWA, and group manager of NIWA’s scientific programming group, with many years of experience in scientific computing and HPC. He helps researchers solve their computational problems around numerical algorithms, code performance, data processing, and output visualisation, with particular focus on weather and climate prediction based on the Unified Model. Wolfgang regularly contributes to projects in the Unified Model Partnership with the UK Met Office, most recently getting involved in code optimisation for its successor model “LFRic”. He is also leading an effort to trial cloud HPC services for operational weather forecasting.

Slides