NCI currently operates two large-scale peak systems, Raijin (a Fujitsu Primergy cluster, commissioned in 2013, and which entered production in June 2013), and Vayu (a Sun/Oracle Constellation cluster, commissioned in 2009, and due to be decommissioned later in 2013), and a small specialised system, Fujin (a Fujitsu PrimeHPC FX10 system, commissioned in March 2013).
Since approximately 2009, NCI has been progressively building infrastructure and services in step with the advancing needs of data-intensive science in a number of research communities. Amongst the early services pioneered by the NCI Cloud and Online Services Team were:
- Services for managing access to, and serving, datasets through specialised interfaces, e.g. the large CMIP5 climate dataset through the Earth System Grid, astronomical data through virtual observatories;
- Virtual laboratories and specialised environments running on virtual machines under OpenStack and VMWare providing tailored services which provide user-friendly interfaces to complex computationally- and data-intensive tasks and workflows—including the ability to spawn tasks onto the HPC system for computationally-intensive tasks, and onto the cloud system for data-intensive processing.
The NCI Specialised Facilities program derives from obligations within the NCRIS NCI contract (2007-11) to establish specialised system resources for targeted application areas.
The specific requirements for such facilities was identified following expressions of interest that were sought nationally, with two facilities being funded, namely the:
- Specialised Facility in Bioinformatics, located at the University of Queensland—with institutional partners CSIRO, UQ, QCIF, QFAB, the State Government of Queensland and NCI, and
- Specialised Facility in Imaging and Bioinformatics (and known now as MASSIVE), located at Monash University—with institutional partners CSIRO, Monash University, the Australian Synchrotron, the State Government of Victoria, VPAC and NCI.
National Environmental Research Data Interoperability Platform (NERDIP) to enable research data carefully prepared, standardised, structured so that it can be used in data-intensive in a rich high performance computational environment.