Friday, June 15th, 2012
Canberra, 15 June 2012 – Fujitsu, a global technology company, announces today that it has been named as the successful bidder to supply a High Performance Supercomputer to the Australian National University (ANU) to be installed within the National Computational Infrastructure (NCI). NCI provides high-end capability computational services to the Australian research community.

As part of this contract Fujitsu will also establish a Collaboration with the ANU in a number of research areas related to strategic objectives common to NCI and Fujitsu.

ANU receives funding for NCI under the Federal Department of Innovation, Industry, Science, Research and Tertiary Education under its Super Science initiative. NCI is also supported by a partnership involving the CSIRO; Bureau of Meteorology; Geoscience Australia; and the ANU. Researcher access to NCI facilities and services is also supported by the ARC and a number of Australia’s research intensive universities.

NCI facilities combine a new data centre, which is due for completion in September 2012, and the Fujitsu supercomputer, due to be fully operational by January 2013.

The NCI supercomputer will take Australian research capacity to new levels in areas such as Climate and Weather, Physics, Astronomy, Geosciences, Chemistry and Advanced Materials.

Fujitsu’s technical solution is to provide the ANU with the specified Petascale performance of 1.2 PetaFlops and 12 Petabytes storage capacity. The new computer will have the power of 56,000 computers working in parallel, and it will have the disk storage of the equivalent of 20,000 computers – but working much faster.

The project will also see the implementation of Fujitsu’s PRIMEHPC FX10 supercomputer, which improves upon Fujitsu’s supercomputer technology employed in the world’s fastest supercomputer, the K Computer. Fujitsu elected to bid its Fujitsu PRIMERGY X-86 High Performance Computing (HPC) technology in order to meet the ANU’s stringent performance, efficiency, and benchmarking requirements and to support the continued use of software codes developed for the X-86 platform.

Fujitsu’s outstanding bid was built on detailed architecture and pricing configurations as well as robust technical and collaborative offers. Its innovative HPC design is based on commodity hardware, which will deliver improved price/performance; access to a greater range of ISV applications; and simplify the migration process of existing X-86 applications.

The ANU supercomputer will be the largest X-86 HPC installation of any brand in the southern hemisphere and the largest PRIMERGY deployment worldwide.

With a theoretical performance of 1.2 PetaFlops – 1,200,000,000,000,000 floating point operations per second – the supercomputer will be an X-86 cluster using Intel Xeon E5 CPUs.

Commenting on the role of the new computer, ANU Vice-Chancellor Professor Ian Young said: “The new supercomputer will provide Australia with a much needed capability to meet national challenges. It will take Australia’s research to new levels in areas such as weather and climate modelling, computational chemistry, particle physics, astronomy, material science, microbiology, nanotechnology and photonics.”

Mike Foster, Chief Executive Officer of Fujitsu Australia and New Zealand said: “Once completed, the NCI Supercomputer will be one of the largest and fastest computers in the world. This win confirms Fujitsu’s ability to leverage its global resources to complement its strong capability in cloud services, application development and managed services at a local level. We are proud to be able to assist the ANU and NCI in playing its vital role as one of the leaders in Australian research.”

The installation will be undertaken by Fujitsu’s combined supercomputing expertise from Australia and Japan with support from Fujitsu Australia engineering teams and sub-contract partners. The Fujitsu project partners are: APC power and cooling; Brocade and Xenon/Mellanox InfiniBand switching; Data Direct Networks high speed storage; and Intel chipsets. Work will commence immediately with the supercomputer to be fully operational by early 2013.