Researchers are now making use of the Australian National University’s new supercomputer, Gadi, one of the most powerful of its kind and the most powerful in the southern hemisphere.
- The artwork featured on Australia’s new supercomputer is by Ngunnawal artist Lynnice Church
- With enhanced processing capabilities, Gadi offers hope for vastly improved research
- The Bureau of Meteorology hopes to use Gadi to better predict bushfire behaviour
Gadi, based at the National Computational Infrastructure (NCI), was unveiled this week after a few months of early operation, described by the university as a “recognised powerhouse in the global supercomputing ranking”.
The supercomputer, which operates 24/7 for 360 days a year, uses the same amount of electricity as it takes to power a medium-sized suburb.
It is also significant for its name, which means ‘to search for’ in the language of the Ngunnawal people, the traditional custodians of the land where the university is located in Canberra.
Supercomputer symbolises a ‘gathering of knowledge’: ANU
The ANU said they chose the name in 2019, which the United Nations General Assembly declared the International Year of Indigenous Languages, to raise awareness of the “crucial role Indigenous languages play in people’s daily lives and culture”.
With a new supercomputer in the works that would become the most powerful in the southern hemisphere — and ranked 25th in the world — they decided to pay tribute to the Ngunnawal tradition of “searching for knowledge in the sky, water, stars and earth all around them”.
The name Gadi was chosen in consultation with the United Ngunnawal Elders Council.
Gadi was also adorned with its own artwork, hand-painted by renowned Ngunnawal artist Lynnice Church.
NCI director Professor Sean Smith said it was a “great honour … to have such a perfectly suited name and artwork as the face of Australia’s new supercomputer”.
“We are proud to be able to make this connection between the traditional owners of the Canberra region and its newest tool of scientific discovery,” he said.
Gadi sits 25th in world ranking
The $70 million project to replace Gadi’s predecessor, Raijin, has lifted the supercomputer’s ranking from number 239 in the world to number 25.
Researchers have already been using the supercomputer for world-leading research since it first became available to some teams late last year, adapting their work to aim even higher.
They include the researchers behind an “industrial-strength” 24-hour build of a 4,012 human genome dataset for studies into cancer, diabetes, and heart disease.
ANU astronomers are also using their time with Gadi to better understand how stars form.
And a team of chemists has been incorporating the supercomputer into its search for a COVID-19 treatment since the pandemic forced them to turn away from their previous work on diabetes.
“We have over 5,000 users that rely on NCI’s unparalleled high-performance computer and data services to produce world-leading scientific research,” Professor Smith said.
“Gadi is, quite simply, a critical foundational infrastructure for Australia and Australian research.”
Supercomputer heralds improved climate forecasting
Among the more than 5,000 individuals using the supercomputer — remotely from their offices — will be researchers from the Bureau of Meteorology, CSIRO and Geoscience Australia.
Both CSIRO and Geoscience Australia will use the supercomputer to improve their systems aimed at predicting extreme weather patterns, including fires, earthquakes, tsunamis and cyclones.
The CSIRO’s executive director of Digital, National Facilities and Collections, Dave Williams, said having an asset like Gadi was “critical” for predicting Australia’s future.
He said its computational models would contribute to the CSIRO’s Decadal Climate Forecasting Project to produce “near-term” climate forecasts.
“The data that Gadi will process and analyse is collected here on Earth as well as by an increasing number of satellites orbiting our planet.”
Modelling could protect communities from bushfire: BOM
The Bureau of Meteorology said they would use Gadi to create fire-atmospheric models in a bid to better understand how they might behave.
They hope its modelling will help them to better predict both the path of a fire and the spread of smoke, by representing the interactions between the heat of the fire, the release of moisture, and atmospheric circulations.
The Bureau’s Science and Innovation Chief Scientist and group executive Gilbert Brunet said the goal was to better protect communities and firefighters.
“Improvements such as those from the fire model and specific event case studies are used to better plan fire suppression operations, reduce risk to firefighters and assist the community in affected areas.”