Publication

Improving the performance of National Centre for Earth Observation (NCEO) code using GPUs

Published: 2023 | DOI: 10.17031/wtr8-2k34

Nicola Martin, Daniel Clewley, Steve Groom

This publication investigates the energy saving impact in running various tools and types of code for GPU acceleration.

Executive summary

The aim of this study was to investigate the advantages of different tools designed for running code on Graphics Processing Units (GPUs) and specify the types of code best suited for GPU acceleration.
Three examples of code were obtained from NCEO scientists for three distinct applications and ported to GPUs on the Natural Environment Research Council (NERC) Earth Observation Data Acquisition and Analysis Service (NEODAAS) MAGEO computing cluster.
Comparisons were made between the time taken to run the code on the original Central Processing Unit (CPU) and the MAGEO CPU and GPU.
Accelerations of between x30 and x1800 were achieved: more details are provided below.
In terms of energy saving this relates to an estimated 93.9% to 99.8% reduction in electricity usage.
The study highlighted the value of expertise in GPUs and coding such as provided by NEODAAS

Download this publication

Please enter your details below to access the publication.

The information you provide will be used to gain a greater understanding of our audience.

Your details will be kept securely and in accordance with our Privacy Policy.

Cite this publication

Martin, N., Clewley, D., Groom, S. (2023). Improving the performance of National Centre for Earth Observation (NCEO) code using GPUs. PML Publishing. DOI: 10.17031/wtr8-2k34