You are here

High Performance Computing (HPC)

What is HPC?

HPC is the use of large-scale, high performance and high-throughput  computers and parallel processing techniques for solving complex computational problems. The terms “HPC” and “supercomputing” are sometimes used interchangeably. HPC machines are often based on a Unix/Linux operating system and involve a batch system and scheduler to maximise the compute utilisation.

HPC systems are most often useful for researchers:  

  • With parallelizable tasks (parallelism can be achieved on a fine or coarse grain level — on a fine level you have a program which is intrinsically parallel, while on a coarse grain level you might run many instances of the same program for parameter studies, or high throughput).

  • With computing jobs needing more memory (RAM) or storage than available on their local systems.

  • With jobs that can make use of accelerators such as GPUs, FPGAs and Phis.

HPC can fit naturally into a researcher’s workflow with links to desktop computers, cloud computers and other compute services. An engineer for example might design graphically a wire frame model on his PC and send this off to an HPC or cloud cluster for optimizations/calculations.

What are Intersect’s HPC Services?

Intersect provides HPC resources to member organisations  and other customers through our partnership with the National Computational Infrastructure (NCI) peak facility in Canberra. Further information on NCI can be found here.   We offer specialised consulting, technical support, training and advice for HPC that augments the services provided by NCI, including:

  • Assisting users to get started, particularly for Higher Degree by Research (HDR) students, including: project applications, user accounts and compute resources.

  • Helping with software compilation and installation.

  • Fixing and optimising Portable Batch System (PBS) scripts.

  • Helping with obtaining appropriate storage and compute quotas including wall-time exemptions.

  • Optimising software applications.

  • Advising on algorithms, software applications and libraries, and software scaling.

  • Profiling of software.

  • Creating automated analysis pipelines.

  • Analysing, designing and implementing very high speed data transfer solutions between NCI and Intersect Space Storage based on Intersect’s 10gbps data link to NCI.

  • Consulting on campus to  support researchers and their collaborations including “hacky-hour” style sessions.

  • Training in HPC from beginner courses to advanced parallel computing courses - refer below.

  • Providing specialist advice for research disciplines including chemistry, physics, mathematics, bioinformatics and engineering.

We also provide the following extended HPC services:

  • Co-ordination of government-funded HPC grant applications.

  • Consulting and advising on supercomputer technical specifications.

  • Project managing your supercomputer setup and installation.

  • Operating your supercomputing facility for you.

  • Providing specialised technical training on supercomputer operations.

  • Formulating and managing your supercomputer migration plan including decommissioning.

Intersect’s HPC Experience

We have a successful track record in setting up and running supercomputers of any scale up to and exceeding 1,000 users, having operated three supercomputers: Barossa, McLaren, and Orange.  Barossa (2003-2009) was a 300 node cluster and the first teraflop machine in academia in NSW. It was in the Top500 (place 113). McLaren (2007-2013) was a shared memory machine and the first terabyte machine in academia in NSW.  Orange (2013-2017) was a very successful 100 node cluster supporting 1,000 researchers from 12 universities in NSW, Vic. From 2017, Intersect operates a partner share of the NCI facility in Canberra.  

Who can access our HPC services?

  • Researchers from Intersect member organisations: HPC resources are merit-allocated and fully subsidised.  

  • Intersect Affiliates and Non-Intersect members: Resources can be purchased, please contact us at help@intersect.org.au.

How do I get access to Intersect HPC Services?

Contact us at  help@intersect.org.au.

If you are interested in Intersect’s Extended HPC Services, please e-mail Joachim@intersect.org.au.     

What HPC training is available?

Intersect offers training courses on HPC topics as follows:

  • Intensive Beginner to Intermediate HPC - Introduction to Unix for HPC: a one-and-a-half day course introduction to the command line and our HPC computing facilities.  A basic familiarity with Windows or Mac OS X is assumed.

  • Intermediate HPC - From Unix to HPC: a 3-hour course introduction to HPC computing facilities for those with Command Line Unix Experience.

  • Advanced HPC - Parallel Programming: An introduction to Parallel Programming in C or Fortran using MPI and OpenMP.  

Tailored courses can also be developed as required.  

Where can I find more detailed information on using Intersect HPC?

Check out Time.intersect.org.au/supercomputing for practical information including:

  • Accessing HPC.

  • Setting up a software environment.

  • Compiling code.

  • Running jobs.

  • Using disk space.

What costs are involved?

  • Intersect Member: Fully subsidised through membership.

  • For non-Intersect members: By commercial negotiation.