Hewlett Packard Enterprise has built a new supercomputer for the Meteorological Service Singapore.
The supercomputer, named SINGV, will enable the service to advance weather forecasting and tropical climate research in Singapore and the broader Southeast Asia region through improved numerical model configurations and data assimilation of ground and space-based weather observations.
The new supercomputer, built using the HPE Cray supercomputer, includes capabilities for sub-kilometer scale urban modeling (uSINGV) to represent urban environments more accurately, as well as coupled ocean-atmosphere-land-wave modeling systems (cSINGV) to understand the relationships between the atmosphere, land, and ocean with weather and climate.
SINGV has a peak performance of 401.4 teraflops, almost 2x the amount MSS previously had access to. The supercomputer is composed of 196 3rd Gen AMD EPYC processors, HPE Slingshot, Cray ClusterStor E1000 storage parallel system from HPE and will run on HPE Cray Operating System and HPE Cray Programming Environment.
SINGV was developed by MSS’ Centre for Climate Research Singapore (CCRS) through collaborations with UK Met Office and partners. The project has been underway since 2013.
Trish Damkroger, chief product officer and senior vice president, HPC, AI, & Labs at HPE, said: “We are honored to play a role in the nation’s digital agenda and be selected by the Meteorological Service Singapore to build them a powerful system with advanced, end-to-end supercomputing technologies that will fuel Singapore’s weather intelligence and speed up predictions of extreme climate events.”
It was announced in April 2021 that HPE-Cray was selected for the UK Met Office’s new supercomputer offering 60 petaflops at peak performance, similarly intended for use as a tool to provide more accurate weather prediction and climate change modeling.