The University of Queensland in Australia Using Dell Technologies HPC

• Built with Dell EMC HPC technology, the Wiener HPC uses Finite Element Method (FEM) to analyse human skull models.
• Research highlights the potential of ultrasound to delay the onset of Alzheimer’s disease.
• The University of Queensland’s GPU-accelerated HPC uses the NVIDIA Volta architecture.

Dell Technologies Inc. announces one of the first computational projects for The University of Queensland‘s Dell Technologies HPC system may enable a non-invasive disease-modifying strategy for Alzheimer’s disease.

The HPC system, built by Dell Technologies for the university’s Research Computing Centre (RCC), is a GPU-accelerated HPC. GPUs, with more cores than CPUs, are suited to processing massive amounts of computational tasks in parallel, including intensive tasks such as data visualisation and ML. It is also used for modeling possible treatments for some of the most debilitating illnesses, such as Alzheimer’s disease.

The Queensland Brain Institute, the university’s neuroscience research institute, is using the Wiener system to model the behaviour of ultrasound using an analysis technique called Finite Element Method (FEM). The modeling calculates what happens to each element of the brain when an ultrasound is passed through the skull. It is hoped that ultrasound can be used to temporarily allow direct delivery of therapeutic drugs to the brain, something not currently possible due to the presence of a blood-brain barrier and activate cells that can digest the plaques that are a hallmark of Alzheimer’s disease.

The promising results will be confirmed in a sheep study, an animal with similar skull properties as humans, and may be instrumental in developing treatments that stop or reserve degeneration, rather than just relieving symptoms.

“Australia prides itself on its research achievements, especially in medicine,” said Chris Kelly, VP, infrastructure solutions group, Dell, AsiaPac and Japan. “With this HPC, the University of Queensland can harness ML to drive innovation, across a broad range of use cases, that previously wasn’t possible. We’re honored to play our part in the resulting discoveries that can change lives for the better.“

Growing to Meet Demand
This research is just one of the Weiner’s many workloads. Demand to implement the HPC in projects across the university has led to the expansion of the system’s power and size, allowing for a broader range of applications, including climate modeling, psychological testing and learning, and disease identification.

“It’s become a whole ecosystem,” said Jake Carroll, CTO, research computing centre, The University of Queensland. “Wiener has become a plethora of massive ML and deep learning capabilities in the organisation. It’s the focal point of AI computing infrastructure at The University of Queensland.“

In other projects, The University of Queensland’s School of IT and electrical engineering is working on developing new digital pathology techniques for faster blood samples results, while another ML algorithm will be able to diagnose the presence of skin cancer from histology slides with the accuracy of a trained pathologist.

The Wiener HPC also supports the university’s lattice light sheet microscope (LLSM), using an image restoration technique called deconvolution to provide clear, real-time 4D biology imaging.

Staying on Leading Edge
The Wiener system, initially built on Dell EMC PowerEdge R740 servers provided a consistent building block for processing data sets across 15 compute and analysis nodes, along with two additional nodes for visualisation.

The system’s expansion, this year, has been driven by demand from multiple departments and research teams – both at the university and with external research partners. The university’s team worked in close collaboration with Dell to engineer and develop the system’s capabilities, using PowerEdge C4140, doubling the initial nodes and tripling the compute performance.

Dell provided a new form factor for accelerated computing using a PowerEdge C4140 platform that does not use a PCIe switch to get information back to the CPUs, resulting in 11% faster training for ML and allowing for lower latency internodal communication for distributed deep learning workloads. Together, these provide the university with measured competitive capability. And, as an early adopter of dedicated NVIDIA Volta-based GPU-accelerated HPC in Australia, it is an advantage the university plans to maintain.

The NVIDIA Volta V100 Tensor Cores gives the Wiener system a total performance of 11.3 petaflops. This gives it the processing power to restore the images at the edge of the university’s network, where the scientific instruments are, providing an interactive experience to researchers viewing 4D data sets in real-time.

“We are seeing exponential growth in both velocity and volume of data being generated by scientific instruments around the world,” said Andrew Underwood, field CTO, HPC and AI, Dell, AsiaPac and Japan. “This is driving Dell and our customers to collaborate on approaches that leverage parallel processing power from our Dell EMC PowerEdge platforms, in order to unlock the confluence of simulation, analytics and ML.“

Video: example of Lattice light sheet imaging of live, ruffling macrophages