KAUST has been awarded the ‘Nobel’ prize of high-performance computing’s top prize - the ACM Gordon Bell Prize for Climate Modelling - in partnership with NSF National Center for Atmospheric Research, USA and other partner institutions.

KAUST was a double finalist this year with two projects and won the ACM Gordon Bell Prize for Climate Modeling for the paper: Boosting Earth System Model Outputs and Saving Petabytes in Their Storage Using Exascale Climate Emulators. This was a collaborative effort from KAUST lead researcher Sameh Abdulah, lead Principal Investigator (PI) Professor Marc G. Genton, Zubair Khalid, Professor David E. Keyes, Hatem Ltaief, Professor Georgiy L. Stenchikov, Associate Professor Ying Sun, and Postdoc Yan Song. The partner institutions were the NSF National Center for Atmospheric Research, the University of Notre Dame, Saint Louis University and NVIDIA.

The Gordon Bell Prize is a prestigious award named after the pioneering computer architect who founded the US National Science Foundation’s Directorate for Computing and Information Sciences. The prize, in its 37th year, is awarded annually to recognize outstanding achievements in the field of high-performance computing (HPC) applied to real-world systems.

This is the first time an institution in the Middle East has won an ACM Gordon Bell Prize, a significant milestone for Saudi Arabia and the region. Lead author Sameh Abdulah commented: “Thank you to the entire team for their dedication and hard work. Winning the HPC Gordon Bell Prize is a significant milestone, not just for us but also for KAUST, reaffirming our position as the leading university in the region.”

“Words cannot describe how excited I am about this once-in-a-lifetime prize, but this is exactly why I decided to join KAUST over a decade ago, to make it happen,” commented Lead PI, Professor Marc G. Genton. Professor Keyes commented, “This is the third consecutive year of being a finalist and it is exciting to finally bring home a Gordon Bell prize for the young researchers on the team and for Saudi Arabia”.

While climate modelling has been a scientific practice since the 1950’s, recently introduced exascale supercomputers (which can process a quintillion calculations each second) offer the opportunity to understand climate change at a far more advanced level than ever before. With the use of exascale computers, computer scientists and climate scientists have developed extremely high-resolution Earth System Models (ESMs).

ESM’s offer great promise in understanding the Earth’s climate, but they require a great deal of computation time and energy, and they require a tremendous amount of storage for the massive quantity of data they generate.

In this paper, KAUST scientists, in collaboration with partners, developed an exascale climate emulator to meet the rising computational and storage demands of high-resolution ESMs.

A statistical emulation of a system, like the global climate, is able to reproduce statistics of interest concerning the system without requiring an ensemble of detailed simulations of the system. It employs a small number of detailed simulations or direct observations of a system to develop a more compact model that can be probed at will for expected behavior of the system at arbitrary points in space or time, such as the number of days in a crop-bearing season in which rain is expected or expected winds at a particular hour for which windfarm capacity is being sized at a given location, or the snow expected in a ski week in a particular resort. Without an emulator, such results would come from averaging simulations or observations over vast volumes of data that would need to be stored.

As demonstrated on multiple supercomputers, including KAUST's Shaheen III, the emulator could save several petabytes of computing storage space. By way of comparison, one petabyte is equal to the storage capacity of approximately 170 top-end servers.

Hatem Ltaief and David Keyes have co-authored all four of KAUST’s Gordon Bell finalist papers over the past three years, along with different applications experts in statistics, climate, genomics and seismic imaging.

The ACM Gordon Bell Prize for Climate Modelling was presented during the International Conference for High Performance Computing, Networking, Storage and Analysis (SC24), was held November 17 – 22 in Atlanta, Georgia.

KAUST is a key player in Cop16 this December in Riyadh, with the University sharing updates regarding its applied climate solutions with the policymakers and leaders in attendance.

The full list of authors are: Sameh Abdulah from KAUST, Allison H. Baker and Informational Science Lab, NSF National Center for Atmosphere Research USA, George Bosilica from NVIDIA, Qinglei Cao from Department of Computer Science, Saint Louis University USA, Stefano Castruccio from Department of Applied ad Computational Mathematics and Statistics, University of Notre Dame USA, Marc G Genton, David E. Keyes, Zubair Khalid, Hatim Ltaief, Yan Song, Georgiy L. Stenchikov and Ying Sun from KAUST.