Common terms used at NCI for high-performance computing, data services and more.

  • ACCESS – The national weather prediction modelling suite, the Australian Community Climate and Earth-System Simulator.
  • Bandwidth – The amount of data that can be transmitted in a fixed amount of time, generally measured in bits per second.
  • Byte – One byte is a typical unit of digital information. It is made up of 8 bits of data, each bit representing a single binary number.
  • Cloud Computing – The on-demand availability of computer system resources, valuable for interactive data processing and analysis using NCI Services.
  • CMIP – The Coupled Model Intercomparison Project, a collaborative framework used by the global community to share coupled models as part of international climate modelling and reporting activity.
  • Cores – A core is a sub-unit of a computer’s Central Processing Unit. Each core is able to work independently on a dedicated task, or share tasks between other cores for efficient parallel processing.
  • Core hour – The amount of computational work performed by one core in one hour.
  • CPU – A Central Processing Unit, the main processor or brain of a computer, which is split into multiple cores capable of working independently. Typical CPUs at NCI have around 18 to 24 cores.
  • DeVL – Data-Enhanced Virtual Laboratory, virtual environments giving researchers easy access to large datasets, analysis tools and computational resources for investigating particular research domains.
  • FAIR data principles – The FAIR data principles state that data should be Findable, Accessible, Interoperable and Reusable, thereby enabling broader use of data within and across research communities.
  • Fat-tree – A universal network structure providing efficient communication across the nodes of a supercomputer.
  • Filesystem – The system used to manage, store and retrieve data files.
  • Flops – Floating Point Operations per Second, a measure of computational performance.
  • Gadi – NCI’s new supercomputer in 2019, Gadi will provide increased performance and capacity to the Australian research capacity.
  • GeoNetwork – NCI’s portal to the data collections, providing download links as well as metadata, author information and access conditions.
  • Giga/Tera/Peta – Unit prefixes denoting factors of a billion (1 followed by 9 zeros), a trillion (1 followed by 12 zeros) and a quadrillion (1 followed by 15 zeros). NCI currently stores over 50 petabytes of data and operates a multi petaflop supercomputer.
  • GPU – Graphics Processing Unit, a specialised processor with distinctive highly parallel structures making them effective at processing large blocks of data in parallel.
  • GSKY – NCI’s scalable, distributed geospatial data server (pronounced ji-skee), which surfaces the large geospatial data stored at NCI using on-demand processing for real-time analysis of nationally significant datasets.
  • HDR – High Data Rate, the 200 gigabit per second generation of Mellanox’s networking communications technology Infiniband.
  • HPC – High-Performance Computing, the use of parallel processing for running advanced scientific and research applications on large computing systems.
  • HPD – High-Performance Data, the use of combined big data collections and high-performance computing for advanced analytics and data processing.
  • Lustre – A parallel distributed file system, used at NCI to manage dozens of petabytes of research data in a powerful, efficient and secure way.
  • NCMAS – The National Computational Merit Allocation Scheme, a merit-based scheme providing researchers from Australian research organisations with access to Australia’s major national computing facilities.
  • NERDIP – The National Environmental Research Data Interoperability Platform, an integrated platform for users to access datasets managed at NCI using access services, data discovery standards and storage infrastructure.
  • Network – The communication system carrying data across the nodes of a High-Performance Computing cluster.
  • Raijin – NCI’s supercomputer from 2013 to 2019, which was ranked 23rd fastest in the world at launch.
  • Service Unit A measure of computational work at NCI.
  • Supercomputer – A supercomputer is a computer made up of thousands of connected compute nodes designed to work in parallel and solve large, complex scientific research and modelling problems.
  • THREDDS – NCI’s primary data server through which data collections are accessed.