McMaster Students Create Fractal Movies Using BlueGene/Q Supercomputer

May 30, 2014 in blog, blog-general, for_educators, for_press, frontpage, in_the_news, news, Uncategorized

superfractals

Computing and software students at McMaster University created some stunning videos of fractals using the BlueGene/Q, one of the most powerful computers in the world, administered by SOSCIP and hosted by SciNet.

Read the full articles on McMaster University’s Daily News’ or on HPC wire.

The videos can be seen on YouTube.

Details on the SCOSCIP BlueGene/Q at SciNet can be found on our wiki.

International Summer School 2014 on HPC Challenges

February 13, 2014 in blog, blog-general, for_educators, for_researchers, for_users, news

intlsummerschool4
Apply by 9 March, decisions in early April
Expenses paid by program
Sponsored by PRACE, XSEDE, Riken, and Compute Canada
website: http://summerschool.niif.hu

Compute Canada/Calcul Canada invites students and researchers at Canadian post-secondary institutions to apply for one of 10 spots allocated to Canada for the fifth International Summer School on HPC Challenges in Computational Sciences. This is a great opportunity for Canadian students and postdocs to attend an Advanced Summer School on High Performance Computing Challenges, all expenses paid.

The workshop is aimed primarily at graduate students or postdocs; however, junior faculty or advanced undergraduates are also welcome to apply. Attendees will be expected to have some experience in HPC parallel programming (for instance, MPI, OpenMP, or CUDA/OpenCL), preferably on software used in successful research projects, and must be at least 18 years of age at time of application. Attendees from all disciplines are invited to participate.

The summer school is sponsored by the European Union Seventh Framework Program’s Partnership for Advanced Computing in Europe Implementation Phase project (PRACE-3IP), U.S. National Science Foundation’s Extreme Science and Engineering Discovery Environment (XSEDE) project, RIKEN Advanced Institute for Computational Science (RIKEN AICS), and Compute Canada / Calcul Canada.

Leading American, Canadian, European and Japanese computational scientists and HPC technologists will offer instruction on a variety of topics, including:

  • Access to EU, Canadian, Japanese and U.S. HPC-infrastructures
  • HPC challenges by discipline (e.g., bioinformatics, computer science, chemistry, and physics)
  • HPC Programming Proficiencies
  • Performance analysis & profiling
  • Algorithmic approaches & numerical libraries
  • Data-intensive computing
  • Scientific visualization

The expense-paid program will benefit advanced scholars from European, Canadian, Japanese and U.S. institutions who use HPC to conduct research. Interested students should apply by March 9, 2014.

Meals, housing, and travel from Canada, Japan and the U.S. will be covered for the selected participants. Applications from students in all science and engineering fields are welcome. Preference will be given to applicants with parallel programming experience, and a research plan that will benefit from the utilization of high performance computing systems.

For further information and to apply online, please click here.

Why Data Centre Providers Love the Greater Toronto Burbs

February 5, 2014 in blog, blog-general, in_the_news, news

The recent announcements of continued IT infrastructure building in Markham (and across the other southern York Region municipalities of Richmond Hill and Vaughan) reflect an established data centre cluster in the area, including recognizable names such as IBM, Rogers, Compugen, OnX, and HP. Of particular noteworthiness is Vaughn-based SCINET—Canada’s largest supercomputer data centre—a High Performance Computing consortium of the University of Toronto and affiliated Ontario hospitals.

Read the full article on TechVibes.

2014 Call for Compute Canada Resource Allocation Proposals

September 24, 2013 in blog, blog-general, for_educators, for_researchers, for_users, news

2014RAC
Any Canadian academic researcher may obtain a default allocation on any Compute Canada system, including those at SciNet, at any time by registering with the Compute Canada DataBase (CCDB) and requesting accounts at one or more consortia. The size of default allocations vary by system.

A Principal Investigator who requires more than the default allocation (be it computing time or storage space), and who is eligible to apply to national granting councils for funding, must submit a proposal to Compute Canada’s Resource Allocation Committee (RAC). The Call for Proposals is posted on the Compute Canada site each fall, with awarded allocations running 1 Jan to 31 Dec of every year.

The 2014 call for resource proposals is now open. Proposals with details about the scientific and technical aspects, are to be submitted via the Compute Canada website CCDB site.

More details can be found in Compute Canada’s Call for Proposal Announcement.

Note: The Resource Allocation deadline has been extended. Proposals must be submitted electronically to Compute Canada on or before October 21, 2013 at 3pm (Eastern). It is however necessary to have started the application process by October 16.

SciNet at the University of Toronto Selects NextIO vCORE™ Express 2070 GPU Computing Solutions

April 5, 2011 in in_the_news, news

Pioneer in rack-level IO consolidation and virtualization solutions at SciNet:

SciNet at the University of Toronto Selects NextIO vCORE™ Express 2070 GPU Computing Solutions

NextIO, the pioneer in rack-level IO consolidation and virtualization solutions, today announced that The SciNet Consortium at the University of Toronto has selected the company’s vCORE Express 2070 GPU systems to power scientific research in a variety of academic areas including astrophysics, aerospace, cosmology simulations and computational combustion.

For more information please visit: MRO Magazine

 

HPCS and ORION in the News

May 21, 2010 in in_the_news, news

HPC community converges at HPCS2010

Supercomputers help Canada address world’s top research challenges

TORONTO – Canada’s supercomputing community, responsible for some of the country’s most exciting scientific discoveries and global research collaborations, will soon converge in Toronto to address how best to support some of the world’s greatest research challenges.
HPCS 2010 (High Performance Computing Symposium), hosted by SciNet and Compute Canada at the University of Toronto June 5 to 9, is Canada’s largest gathering of HPC researchers.
One of the goals of the conference is to raise awareness of HPC and the researchers who use these technologies to address some of the most important research “grand challenges” of our times, from finding the cure for cancer, to deciphering the mysteries of the universe.
“It’s important to appreciate the benefits of Canada’s HPC-enabled research and scientific collaborations. They have a direct relationship to our ability to compete and innovate and to create and sustain jobs in our new economy,” says Chris Loken, Chief Technical Officer of SciNet, Canada’s fastest supercomputer and one of the world’s top HPC facilities.
ORION, which provides the enabling high-speed network infrastructure which allows some 30 Ontario institutions to engage in shared and distributed HPC resources and collaborations, is a major sponsor of HPCS 2010.
“Canada is making significant contributions to global research,” says Phil Baker, President and CEO of ORION. “You can say that our HPC resources are helping Canada punch above its weight when it comes to significant and meaningful scientific discoveries, especially in the areas of the environment and climate change, physics and cancer research,” he says.
Across Canada and around the world, research is increasingly dependent on access to high performance computing, dedicated high-speed research networks and data resources and tools. In Canada, these integrated resources represent close to a petaflop of computing capability and online and long term storage, with access and retrieval over Canada’s national, provincial and territorial high-performance networks.
Suzanne Fortier, president of the Natural Sciences and Engineering Research Council of Canada (NSERC) is among the keynotes addressing the conference, which is co-presented by Compute Canada, the national organization that is implementing a powerful national HPC platform for research with seven partner consortia across the country.
Attending are research leaders from the consortia, including ACEnet (Atlantic Computational Excellence Network), CLUMEQ (Consortium Laval, Université du Québec, McGill and Eastern Québec), RQCHP (Réseau québécois de calcul de haute performance), HPCVL (High Performance Computing Virtual Laboratory – Queen’s University), SciNet (University of Toronto); SHARCNET (Shared Hierarchical Academic Research Computing Network – Ontario); and WestGrid (Western Canada Research Grid).
Delegates will include computational researchers from all disciplines in industry and academia, computer scientists and vendors, including IBM, Intel, Dell and Microsoft among others.
Canada’s extended research and education community, as well as partners from business and industry who want to learn more about the current state and future direction of HPC technologies and their capabilities, are invited to attend the conference.

Learn more at www.hpcs.ca  / www.computecanada.org / www.orion.on.ca

HPCS 2010 in Toronto

November 23, 2009 in news

SciNet will host the 24th annual High Performance Computing Symposium in 2010

HPCS 2010 will be hosted by SciNet in Toronto.  June 5-9, 2010
Workshops: June 5-6
Welcome Reception: June 6
Conference: June 7-9
www.hpcs2010.org

SciNet Director is the first Canadian recipient of the Bower Award

October 20, 2009 in news

University Professor Richard Peltier is the first Canadian to win prestigious international science prize
University Professor Richard Peltier of physics has been chosen by the Franklin Institute to receive the 2010 Bower Award and Prize for Achievement in Science. Peltier is the first Canadian to receive the $250,000 award and joins a group of previous recipients that includes Marie Curie, Thomas Edison, Albert Einstein and Stephen Hawking.

To visit the Franklin Institute website click here.

 

The Bulletin: University of Toronto supercomputer among 20 fastest machines in the world

June 23, 2009 in in_the_news, news

Link to the Bulletin article on SciNet

Tuesday June 23rd 2009
http://www.news.utoronto.ca/media-releases/university-of-toronto/university-of-toronto-supercomputer-among-20-fastest-machines-in-the-world.html

University of Toronto supercomputer among 20 fastest machines in the world

Canada’s fastest and most powerful supercomputer, located at the University of Toronto and built by IBM, has been ranked 16th on the top 500 list of the most powerful commercially available computer systems in the world. The top 500 list was announced today at the annual International Supercomputing Conference in Hamburg, Germany.

“With a high-performance computer that is among the top 20 on the planet, the University of Toronto has become one of the world’s premier computational research institutions,” said Professor Paul Young, vice-president (research) “The supercomputer will attract researchers from around the world and will enable us to conduct research in a variety of disciplines that will have a direct and positive impact on society.”

Designed by IBM and SciNet, a consortium that includes the University of Toronto and affiliated research hospitals, the IBM System x iDataPlex is part of a supercomputer facility that pioneers an innovative hybrid design containing two systems that can work together or independently, connected to a massive five petabyte storage complex able to store the contents of more than one million DVDs – which would stack more than twice the height of the CN Tower. The iDataPlex server is specifically designed for data centres that require high performance, yet are constrained on physical space, power and cooling infrastructure.

With more compute power than existed on Earth at the end of 2001, the iDataPlex is capable of performing over 300 trillion calculations per second – more than 10 times faster than any other research system in Canada. “If you gave a calculator to every single man, woman and child in Canada and had them each do one calculation per second, it would take the entire country four months to do what this machine does every second,” said Chris Loken, chief technical officer at SciNet. It is the world’s fourth most powerful system built on a university campus and the second most powerful university system outside the United States. The extraordinary speed of the iDataPlex comes from the 7,500 Intel Xeon© Processor 5500 series 2.53 GHz processors inside. This model outperforms the previous generation Intel Xeon© Processor 5400 series by over 2.25 times within a similar power configuration.

“Congratulations to the University of Toronto’s SciNet installation being recognized as the largest and most powerful supercomputer in Canada,” said Richard Dracott, Intel’s general manager of high performance computing. “We are honoured to have platforms based on the Intel Xeon© processor 5500 series to drive Scinet’s objectives to solve some of the most complex challenges facing our planet.”

The iDataPlex system will enhance U of T’s competitive position in global research projects in aerospace, astrophysics, bioinformatics, chemical physics, medical imaging and climate change prediction by analysing high-resolution global models to predict future risks, including the faster-melting Artic sea ice. Another immediate project is the construction of regional climate change predictions for Ontario and the Great Lakes watershed region.

The system will also play a key role in the global ATLAS project, using data collected by the Large Hadron Collider in Geneva to explore the modern scientific mystery of why matter has mass and what constitutes the mass of the universe. Because of its hybrid design, the system is extremely flexible and capable of running a wide range of software at a high level of performance.

The system provides up to five times the density versus competitive offerings and can regulate the data centre’s temperature 70 per cent more efficiently with IBM’s Rear Door Heat Exchanger and is entirely water cooled. Intel technologies like Integrated Power Gates, which automatically turn off any processors when not in use, will further assist in energy savings, making the system among the most efficient. The state-of-the-art design saves enough energy to power more than 700 homes for a year.

“We are particularly proud of the cooling system, which makes this one of the most energy-efficient supercomputer installations in the world,” Loken said. “Typically, 40 to 70 per cent of the energy going into a data centre is used for cooling or is lost for other reasons, while we expect our non-computing overhead to be reduced to 16 per cent. The data centre’s design makes use of free cooling for roughly one-third of the year.”

The iDataPlex system adds to SciNet’s existing supercomputer capability, which includes an IBM Power 575 system with 3,328 POWER6 processing cores with peak performance of more than 60 trillion calculations per second. SciNet is one of seven consortia that comprise Compute/Calcul Canada, a national high-performance computing resource for academic institutions. Funding for SciNet has been provided by the Canada Foundation for Innovation’s National Platforms Fund, in partnership with the Government of Ontario and the University of Toronto.

TOP500 Ranking: Net’s GPC ranks higher than previously forecasted

June 23, 2009 in in_the_news, news

The SciNet GPC- iDataplex system, which had been forecasted to rank in the top 25, comes in at a staggering #16 on the TOP500 List of the world’s top supercomputers.  The TCS- Power6 system, which was ranked at #53 in the November list still stays in the running at #80 this summer.