The Trusted Leader in High Performance Computing


CSIRO Deploys SGI UV 3000 for Pioneering Scientific Advancements

April 5, 2016

The Commonwealth Scientific and Industrial Research Organization (CSIRO), Australia's national science agency, selected SGI® to power its scientific research in a number of cutting-edge fields. CSIRO installed the SGI® UV™ 3000 high performance computing (HPC) system, named "Ruby," to handle massive data workloads and enable researchers across Australia to partner on projects designed to help solve complex issues and expand the frontier of scientific knowledge.

Superstar. Supercomputer.

Ruby is CSIRO's fourth SGI system since the organization's first SGI purchase in 2003—an SGI® Altix™ 3700 system named "Cherax." CSIRO's current upgrade to the SGI UV 3000, in-memory computing system, increases the bandwidth for CSIRO's researchers to collaborate on a range of topics including bioinformatics, genome assembly, climate modeling, nanoscience and image and image processing.

Next-gen Sequencing for Next Generation Breakthroughs

The modular design of the SGI UV 3000 enables extreme scalability, designed for compute-intensive workloads such as next-generation sequencing (NGS) and scientific simulations. As CSIRO's research expands across disciplines, with more than 1,200 researchers using the system, the organization needed a more powerful system to tackle a diverse range of workloads.

"Our researchers benefit from an SGI supercomputer because it is able to process a huge amount of data simultaneously," said Denis Bauer, a CSIRO health and biosecurity scientist. "In the case of my area of genomics, Ruby is able to load an entire genome at once. This is a tremendous benefit in helping identify the cause of disease. Ruby is the backbone to making this new reality cost-effective and more accessible than ever before."

Complex Made Simple

Using the latest SGI system, CSIRO researchers can now prepare and manage their data and computing needs throughout the project lifecycle, and collaborate across the organization on various projects. In addition to performing complex calculations, Ruby runs CSIRO's Data Migration Facility (DMF), which can store more than 14 petabytes of information. DMF uses hierarchical storage management to automatically move data between different storage systems. As a result, researchers gain efficiency as devices of different speeds, sizes and costs behave as one unified drive.

Game of Thrones Anyone?

CSIRO discussed in their recent blog post that, "We've been operating our DMF on various machines since 1991, which makes it the longest continually operating DMF deployment in the world. Since it was set up it's expanded by five orders of magnitude to now store more than 14 petabytes of information. To put that into perspective, that's enough to store every episode of Game of Thrones (in HD) 325,581 times!"

Ruby is based on the Intel Xeon processor E5-4627v3 which delivers of 640 cores and eight terabytes of memory using SGI NUMAlink® 6, which connects all 640 cores to appear as one personal computer. SGI NUMAlink delivers extreme scale with single system simplicity. This makes the SGI UV with NUMAlink the perfect system for data-intensive workloads.

"The SGI UV 3000 is designed for compute-intensive, fast algorithm workloads such as scientific simulations," said Gabriel Broner, vice president and general manager of high performance computing, SGI. "SGI is proud to work with CSIRO. Our system gives them application performance, ease of use, and energy efficiency to accelerate their time to discovery and reach the forefront of research."

As the deluge of data increases the complexity of scientific research, organizations like CSIRO require increasingly powerful computing capabilities to handle these enormous workloads. The various SGI HPC solutions enable researchers to prepare and manage several compute-intensive projects simultaneously, providing energy efficient and flexible systems. All this performance capability fits into racks and pods in temperature controlled rooms with other computer hardware.

Supercomputers, particularly SGI machines like Ruby, play an important role in science and engineering—fields becoming increasingly critical to so many aspects of our lives.

Read more on the CSIRO blog.

© Silicon Graphics International Corp. All rights reserved. SGI, the SGI logo, SGI DMF, SGI UV and SGI ICE are trademarks or registered trademarks of Silicon Graphics International Corp. or its subsidiaries in the United States and/or other countries. Intel and Xeon are trademarks or registered trademarks of Intel Corporation. All other product and service names mentioned are the trademarks of their respective companies.