External Resources
In addition to high peformance computing resources managed by the RCC, UQ researchers can gain access to a number of national computational resources. These are:
National Computational Infrastructure (NCI)
The federally-funded National Computational Infrastructure (NCI) operates two large-scale peak systems:
- Raijin (a Fujitsu Primergy cluster)
- Vayu ( a Sun/Oracle Constellation cluster).
NCI also manages a small specialised system, Fujin (a Fujitsu Prime HPC FX10 system).
A key characteristic of both of the production systems, Vayu and Raijin, is their balance.
System balance is an important characteristic that provides hardware and applications scalability by requiring that crucial system parameters such as performance of memory, the interconnect fabric, and the file system all scale with node and aggregate processor performance.
Such balance, which is displayed in the hardware solution architected by the HPC systems team at NCI, provides the facility with strong and scalable performance, and delivers a quality user experience.
15% of NCI's capacity is available for competitive access via the NCI's National Computational Merit Allocation Scheme.
UQ has an additional 7.5% available to UQ researchers, and further capacity (around 4%) can be obtained from the Queensland Cyber infrastructure Foundation.
Nectar
QCIF operates the Queensland node of both the Research Data Storage Infrastructure (RDSI) project (Data Collections) and the national research cloud provided by the National eResearch Collaboration Tools and Resources (Nectar) (Virtual Machines) Project. The Queensland node is called QRIScloud, and its main data centre is currently located at The University of Queensland’s St Lucia campus.
QRIScloud is an integral part of new, national research infrastructure, leveraging data collections stored in the local and national RDSI nodes, and integrating with access to Queensland-based HPC facilities and specialised cloud data services.
Computationally, QRIScloud currently offers about 4,000 cores. UQ researchers can apply for access to a number of virtual machines through QCIF.
MASSIVE
The Multi-modal Australian ScienceS Imaging and Visualisation Environment (MASSIVE) will be the primary Australian high performance computing facility for computational imaging and visualisation.
The MASSIVE facility provides the hardware, software and expertise to help scientists apply advanced imaging and visualisation techniques across a wide range of scientific fields and provides scientists an unprecedented view of captured data and simulated models by providing the capability to view full resolution datasets.
MASSIVE has been formed by the Australian Synchrotron, CSIRO, Monash University and VPAC, with funding from the State Government of Victoria and the National Computational Infrastructure (NCI).
MASSIVE comprises two separate installations, one at the Synchrotron and the other at Monash University. These two interconnected computers operate at more than 5 and 30 teraflops, using traditional CPUs and GPUs.
15% of MASSIVE is available for competitive access via the National Computational Merit Allocation Scheme.
VLSCI
The Victorian Life Sciences Computation Initiative (VLSCI) is a supercomputing centre for life science computations, and predominately serves the Victorian research community.
VLSCI equipment includes:
- a 65,535 core IBM Blue Gene/Q system (Avoca)
- a 1088 core SGI Altix XE Cluster (Bruce)
- a 688 core IBM iDataplex x86 system (Merri)
- and a 1120 core IBM iDataplex x86 system (Barcoo).
Whilst the VLSCI Resource Allocation Committee mostly serves Victorian researchers, 15% of the resource is available for national allocation under the National Computational Merit Allocation Scheme.