SGI Provides Storage at Australian Pawsey Supercomputing Centre

Silicon Graphics
International Corporation announced that iVEC and the Commonwealth
Scientific and Industrial Research Organisation (CSIRO) have selected SGI
to provide the massive data management infrastructure at the Pawsey Supercomputing Centre.

F

The centre is part of the Australian Government Super Science Initiative to
support the Australian
Square Kilometre Array Pathfinder (ASKAP) and the Murchison Widefield Array (MWA) radio
astronomy facility.

The Pawsey Centre will process huge volumes of data. The two largest generators
of data are expected to be the Australian Square Kilometre Array Pathfinder
(ASKAP), Australia’s largest and most capable radio telescope ever constructed,
and the Murchison Widefield Array (MWA) project, which studies the signals from
the dynamic radio sky as well as measurements of the Sun and hemispheric
plasma.

These initiatives will expand understanding of the universe and drive
technological development worldwide. It is anticipated that these two projects
combined will generate eight petabytes of data each year, all of which will flow
through the Pawsey Centre. When adding in data from other research areas, such
as geothermal modeling and rock characterization, iVEC forecasts supporting
data volumes at approximately ten petabytes annually for the foreseeable future. To
manage such volumes, CSIRO selected an SGI InfiniteStorage and SGI UV 2000
based solution to address the scale and cost-efficiency requirements for a
project of this magnitude.

The SGI InfiniteStorage solution comprises disk
storage systems and licenses to support up to 100PBs of online storage that is
virtualized across multiple performance tiers by SGI DMF software, with data
ingest and workflow managed by SGI LiveArc. The system’s 6PB of primary
storage is virtualized to provide further flexibility and savings with an
additional cache of SGI MAID, the company’s energy-efficient zero-watt disk
technology. This environment integrates with 40PB of data tape libraries and
provides expansion capabilities to support a 100PB HSM online
environment.

Beyond big data management and storage, the SGI UV
2000 solution delivers both big data analysis and visualization capabilities.
Working as a set of data-analysis engines to move and process huge amounts of
data very quickly, the UV 2000 can be incorporated into an array of
different workflows to provide pre- and post- processing for a range of
scientific applications. The versatility of the UV 2000 allows the Pawsey
Centre to deliver powerful big data visualization
capabilities, enabling its researchers to view and manipulate vast amounts of
data in new ways. This new technology will allow the Pawsey scientists to
visualize images in the order of 4TB, an order of magnitude greater than
the size of images previously available, which in turn provides a quicker path
to results and interactions.

"iVEC is committed to ensuring
Australia maintains its place as a world leader in research and scientific
computing, and the Pawsey Centre is a critical pillar in this strategy,"
says Neil Stringfellow, iVEC’s executive director. "SGI’s storage and data analysis infrastructure is a vital component of
the Pawsey infrastructure. In particular the SGI UV 2000 visualisation system
with its very large shared memory capability will enable our researchers to
manipulate their data in a completely new way, leading to the potential for new
insight and ambitious analysis."

The Pawsey Centre is expected to be operational and in production
by October 2013, with users gaining early access before June. It is anticipated
that the first component to be heavily used will be the InfiniteStorage
based HSM as the radio-astronomy community moves into the final testing phase
for their apparatus and begins to stream significant volumes of data. The
UV 2000 based data-analysis engines and visualization systems are expected to
come online shortly thereafter.

"For
decades, SGI has been solving big data challenges for researchers across
science and industry in an effort to find answers to the world’s toughest
challenges," said Jorge Titinger, president and CEO, SGI. "We are very pleased to support the data
management needs of the Pawsey Supercomputing Centre. They are conducting
impressive research, and with our InfiniteStorage and UV 2000 technology, will
be able to reach results and interactions more quickly. We look forward to
continuing this partnership and seeing the Pawsey Centre’s revolutionary
solutions to challenges in science."

Several aspects to big data including volume, velocity and variety are
highlighted in a video
featuring Jorge Titinger, president and CEO, SGI.

Technical notes:

SGI InfiniteStorage Environment:

• DMF – Virtualization software and licenses to support expansion up to
  100 PBs of tiered storage.
• CXFS – High performance clustered file system.
• LiveArc – Metadata and data workflow management software.
• 6PB of high performance IS5600 storage
• Nearly 1PB of SGI MAID, power-efficient zero-watt disk, as a low-latency cache.

Data Processing and Visualisation:

• The InfiniteStorage environment is connected to the data
  processing and visualisation systems with two 56GB/s IB networks. An UV
  2000 combined with 34 SGI Rackable C2108 servers underpins the visualisation
  solution and provides a general-purpose virtualised server environment. The
  UV 2000 built with the latest Intel Xeon processors and 6TB of main memory to
  provide high speed in-memory data manipulation. It is augmented with four
  Nvidia Tesla K20 accelerators.
• Two UV 20 servers provide a foundation for LiveArc workflow and
  metadata management software to deliver the complex data workflows within the
  storage and visualisation systems.
• A dozen SGI C2108 servers implement the CXFS clustered filesystem, the DMF
  hierarchical storage manager, and provide gateways between the storage and the
  Petascale and Real-Time computers. Four additional C2108 servers function as
  parallel data movers to transfer massive amounts of data between the disk
  storage and the robotic tape libraries.
• Installation and support provided by SGI professional services.