High Performance Computing
High Performance Computing (HPC) allows the analysis of large or complex problems that are not solvable with laptop and desktop computers.
HPC refers to the practice of aggregation of computing power in a way that delivers much higher performance than one could get out of a typical computer or workstation. This allows for the analysis of very large data sets and the solving of complex problems in areas such science, engineering, health and medicine, or business and marketing.
The terms High Performance Computing and Supercomputing are used interchangeably.
Graphics Processing Units (GPUs) are a specialized type of processing unit that can support machine learning by providing computational power and parallel processing capabilities. Access to GPUs is provided alongside some existing HPC and cloud computing facilities.
What HPC is available?
HPC resources are available for La Trobe’s researchers from national providers and from local clusters on LTU premises.
- National Teir-1 HPC Providers
- Local HPC
See the sections below for more information about each option.
What HPC is right for me?
There are several factors that go into working out which HPC facility you should use. Often this comes down to whether the software you need is already available on one of the machines. Other considerations are how much memory and how many cores you will need, and what type of processing unit is most appropriate for your computation. We recommend contacting our Digital Research Analyst to discuss your needs and the options available.
National supercomputing and cloud computing facilities provide access to HPC and GPU resources. These are primarily made available via existing merit allocation schemes.
The National Computational Merit Allocation Scheme (NCMAS)
NCMAS provides merit-based access to Australia’s two Tier-1 HPC facilities in Australia, the National Computational Infrastructure (NCI) and the Pawsey Supercomputing Research Centre. The application process usually opens in August for the following year’s allocations. Use the button below, or contact Digital Research, if you would like to learn about the NCMAS application process.
Gadi Supercomputer
Gadi, managed by NCI, is the one of the most powerful research supercomputers in Australia. It is CPU-based (no GPUs) and best suited for the largest parallel computational problems. Gadi allocations are available via the NCMAS. However, La Trobe has a small, purchased allocation of resources available for use by our researchers. Contact our Digital Research Analyst to request access to Gadi.
La Trobe has a limited supply of HPC resources on premises. These include the LTU-HPC, ex-Spartan, and Atto Linux Server clusters.
- LTU-HPC: the primary on premises HPC, with 96 cores and some GPU support.
- Ex-Spartan: an older cluster with a small number of cores and limited GPU support.
- Atto Server: a Linux server with 72 cores.
To request access to local HPC or to find out whether these resources can meet your compute requirements, please contact Information Services via the button below.