Page tree

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

NCI supports AI applications on diverse scientific problems such as solving differential equations, accelerating molecular dynamics, predicting 3-dimensional protein structures, controlling nuclear fusion, or weather forecasting with higher resolution and accuracy. Some HPC-AI applications encounter a range of workflow bottlenecks. The performance can be limited by processes that are compute bound, I/O bound, memory bound, or a combination of the above, pushing the need for HPCD into the exascale range. The unprecedented representational ability of deep neural network structures can only be brought to bear on the most complex and urgent global problems by model training with ultra-large datasets on heterogeneous exascale HPCD architectures.Through this program, we will identify implementation hurdles and barriers, provide support to address them for the next generation of young researchers who in turn will empower their future organisations for the broader benefit of the Australian research and industry sectors. 


Value

$10,000 and 100KSU with appropriate storage per annum  2023

Eligibility

...