New Supercomputer Brings Deep Learning Capabilities to CSIRO

The CSIRO has welcomed a new supercomputer to its Canberra campus, with Dell EMC sending the new Bracewell system live earlier this month. The CSIRO put a call out to the market late last year for a new $4 million, 1,87 petaflop supercomputer that would replace its existing BRAGG high performance compute cluster. It contains 1,634,304 Cuda compute cores, 3192 Xeon compute cores, and 29TB of RAM, and runs both Linux and Windows. Australian digital research body Data61’s Computer Vision Group is one of the first to use it. The team is developing software for a bionic vision solution that aims to restore sight for those with profound vision loss, through new computer vision processing that uses large scale image datasets to optimise and learn more effective processing.
"When we conducted our first human trial, participants had to be fully supervised and were mostly limited to the laboratory, but for our next trial we're aiming to get participantssy out of the lab and into the real world, controlling the whole system themselves," said Barnes (Associate Professor).
The CSIRO's deputy CIO Angus Macoustra called the supercomputer a "critical enabler" for the organisation's science, engineering, and innovation activities.

"The power of this new system is that it allows our researchers to tackle challenging workloads and ultimately enable CSIRO research to solve real-world issues," he said.
The upgrades from Dell EMC now sees Pearcey comprise 349 PowerEdge M630 compute nodes, with the additional 119 boasting dual Intel Xeon 10 core CPUs, 128GB RAM, and an FDR InfiniBand network connection that will move data across the supercomputer at 7GB/s per node with ultra-low latency. In addition to boosting the bionic vision work, the system will also provide computational support for a number of science and engineering efforts at CSIRO, including research in virtual screening for therapeutic treatments, traffic and logistics optimization, modeling of new material structures and compositions, and machine learning for image recognition and pattern analysis. With a budget of AU$1.5 million, the CSIRO specified the new ATC was to meet the needs of the radio astronomy research community and high-end researchers in other areas of computational science, such as geosciences, nanotechnology, and biotechnology.