XENON Systems to build research HPC for UQ
The University of Queensland has contracted XENON Systems to build the first-of-its-kind FlashLite supercomputer.
XENON has won the tender to build the FlashLite high-performance computer (HPC) and preconfigure the operating systems and cluster middleware.
The build will be conducted at XENON’s Melbourne headquarters and the FlashLite is scheduled to be up and running at the University of Queensland in July.
The FlashLite was designed by Professor David Abramson, director of the University of Queensland’s Research Computing Centre.
The computer was inspired by the US National Science Foundation’s HPC Gordon. It includes three main innovations - high-speed flash memory, large amounts of high-speed main memory and software shared memory.
Compute nodes in the system can be aggregated together into larger supernodes using the ScaleMP software suite, which combines multiple physically separate servers into one single virtual high-end machine.
FlashLite is being funded by the Australian Research Council in conjunction with CSIRO, Griffith University, Monash University, Queensland Cyber Infrastructure Foundation, Queensland University of Technology, The University of Queensland and The University of Technology, Sydney.
Researchers plan to put the machine to work helping to deliver breakthroughs in research into areas including cardiac disease, climate change and astrophysics.
“FlashLite will solve a problem that is everywhere these days: big data, and how to exploit it in critical research,” Abramson said.“The difference is how we approached the problem; we asked whether existing computing systems were fundamentally built the right way to leverage big data.”
The hidden legal risk in your AI workflow
Australia's new privacy tort just made your AI workflow a legal liability.
Secure by Design: vital in an evolving threat landscape
By embedding security into the fabric of development, Australian business can not only comply...
Keeping IT teams resilient by reducing complexity
Architectural diversity, API sprawl and fragmented tooling make it difficult for IT teams to see...