Press Releases

Planning for HPC Summer School 2015 is underway. Summer School (west) which will be held at Western University, London, Ontario from May 25-29th. This five day summer school includes lectures and labs on programming distributed and multicore systems and various topics pertaining to scientific computing. Attendees will learn how to program distributed memory systems--networked computers--known as “clusters”, shared memory multicore systems and GPUs. Attendees will also have a chance to have a review on the basics of computing and techniques as well as software packages for applications in different disciplines.

A long-running event for SHARCNET, Summer School was introduced to users over a decade ago. Summer School is now an Ontario-wide event which is organized in partnership by the three Ontario HPC consortia: SHARCNET, SciNet and HPCVL. Summer School (central) will run from July 13-17 at the University of Toronto, and Queen’s University will run Summer School (east) from July 27-31. Note that the courses offered by the three sites may be slightly different. Please check the site specific programmes for details.

While there is no fee to attend Summer School, registration is required.

SHARCNET is issuing a new call for proposals for Dedicated Programming Support. This programme provides support for computational projects of exceptional potential that will have lasting impact and value and that require significant support from SHARCNET to proceed.

For Round VI, applications are encouraged that satisfy the programme objectives and priority will be given to proposals that:

• Propose to make novel, effective and large-scale use of non-standard architectures especially GP-GPU.
• Propose to develop innovative visualization applications and techniques that emphasize the visualization of large datasets especially using distributed/parallel visualization/rendering techniques.
• Propose a programme of work that will enable innovative projects from disciplines that are traditionally not major users of HPC.

Applications are submitted via SHARCNET’s webportal and are due by April 26, 2015. Note that users must have a SHARCNET webportal account in order to access the online form.

For additional information, please refer to the application guidelines. Questions should be addressed to research-support@sharcnet.ca.

Compute Canada and its regional partners (ACENET, Calcul Québec, Compute Ontario and WestGrid) are pleased to announce the institutions nominated to host national advanced research computing (ARC) systems in Compute Canada’s next round of infrastructure funding.

The following institutions have been nominated:

• LP Site – University of Toronto
• GP1 Site – University of Victoria
• GP2 Site – Simon Fraser University
• GP3 Site – University of Waterloo

The new systems are planned to be fully operational in 2016. The Canada Foundation for Innovation (CFI) provides significant capital funding for these systems through its Cyberinfrastructure Initiative, in which “Challenge 2” targets the needs of Compute Canada, divided into two stages, one to be decided in 2015, and the next to be decided in 2016. Compute Canada’s operations and management costs are funded through the CFI’s Major Science Initiative (MSI) program, which funds Compute Canada through March 2017. Both of these initiatives (Cyberinfrastructure and MSI) are funded jointly by the CFI, provinces and institutions.

With a total project cost of up to $37.5M, Challenge 2 Stage 1 funding targets the most “pressing and urgent” ARC needs of Canadian researchers. The majority of current Compute Canada systems are already at or beyond their nominal 5-year lifespan. Aging systems lead to increased operational costs due to less energy efficiency, increased likelihood of system failure and increased warranty costs on key components. These systems are not as well-adapted to modern scientific usage as the new systems will be. As such, the focus of this renewal funding is on replacement of existing capacity and to “address the most pressing immediate needs.”

For more information on Compute Canada’s technical plan, a summary document is available. More info

Chris Eliasmith, University of Waterloo Professor in the Departments of Philosophy, Systems Design Engineering and Computer Science and Canada Research Chair in Theoretical Neuroscience wins the NSERC John C. Polanyi Award that honours an individual or team whose Canadian-based research has led to a recent outstanding advance in the natural sciences or engineering.

Eliasmith has built a computer model of the human brain that makes human-like mistakes, has human-like accuracy, and takes human-like lengths of time to process information. The work could lead to better treatments for brain trauma and Alzheimer’s, as well as advances in artificial intelligence. Its name is Spaun, and it’s more human-like than any computer today. Developed by University of Waterloo neuroscientist Eliasmith, Spaun(Semantic Pointer Architecture Unified Network) is the world’s largest simulation of a functioning brain. And unlike other computer brains, Spauncan mimic the human brain’s ability to see, remember and act.

This internationally acclaimed computer program’s 2.5 million virtual neurons and simulated eye and arm allow it to shift between diverse tasks – from copying human handwriting to finding hidden patterns in a list of numbers. Such tasks will help researchers understand how millions of neurons cooperate to cause behaviour.

Eliasmith has drawn on his experience in philosophy, neuroscience, systems design engineering and computer science to develop mathematical theories of the brain that will make it possible for scientists to study the behavioural consequences of brain damage in a safe, simulated environment, without damaging a real brain. It will provide new insights into how the brain actually works and potentially revolutionize the way we treat brain disorders.

Eliasmith’s work has been featured in the BBC, Popular Science, CBC, Wired, New York Times, Science News, Discovery, and Nature just to name a few. He’s also the author of a step-by-step guide, called ‘How to Build a Brain’, which teaches readers how to build their own computer model of the human brain. More info

The sixth International Summer School on HPC Challenges in Computational Sciences will be held from June 21-26, 2015, in Toronto, Canada. This is an advanced summer school on High Performance Computing which targets graduate students and postdocs who already have some experience in HPC parallel programming (for instance, MPI, OpenMP, or CUDA/OpenCL), preferably on software used in successful research projects.

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

Participation in the summer school is decided through an application process. Meals, housing, and travel 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.

Applications are due by March 11, 2015. For further information and to apply online, please click here.