Backgrounder: Supercomputing and SciNet

What is supercomputing?

Supercomputing, or High Performance Computing, is using the largest computers available to tackle the biggest problems facing science, society, or industry. Whether generating computer models of unprecedented fidelity in the aerospace, biological or climate sciences, or analyzing vast amounts of data from astronomy, business data in industry, or textual or musical data in the humanities, supercomputing is a rapidly growing set of hardware and skills applicable to building Canada’s skills and capabilities in science, technology, and the economy.

The computers at SciNet can do in an hour what would take a single PC years to complete, if it were even possible to get the problem to `fit’ on a single computer. This does more than shorten the time to get an answer; it makes new types of analysis and understanding possible.

Supercomputing and the Digital Economy

Supercomputing is one of the essential pillars of the digital economy; it is responsible more than any other level of computing for driving innovation in fields like aerospace engineering, biomedicine, climate science, physics, and astronomy. It is also essential for building digital skills; the students and researchers who begin their careers in supercomputing have skills then immediately translatable to not only other supercomputing-powered research fields, but to the sort of technologies that run Facebook, YouTube, Google, and other new digital media sites relying on huge databases and massive server farms. Supercomputers and the networks that create them are a vital piece of digital economy infrastructure and essential for building a Canadian ICT presence.

Supercomputing and competitiveness

The US knows that supercomputing is vital for economic competitiveness. The US Council on Competitiveness, founded in 1986, is a non-partisan board that includes CEOs and academics whose primary goal is to “…America’s long-term competitiveness challenges.” One of its primary initiatives is to encourage adoption of supercomputing to cost-effectively turbo-charge US firms’ R&D efforts. Most remarkable, perhaps, is its list of success stories as they include not only `high tech’ industries like medical research, drug discovery, aerospace engineering, auto industry virtual crash testing; but also companies producing golf clubs (Ping), tires (Goodyear), plastics and composites (Alcoa), home appliances (Whirlpool), and even seemingly mundane products like shampoo.

The power of computing is that provides a `laboratory’ that can be used by virtually every knowledge worker, whether academic or industrial – and further, they can share the same laboratory, exchanging knowledge and techniques, and providing industry a fertile training and recruiting ground for the next generation of innovators. This is just one way that the digital economy can spur the industrial economy. Partly because of this, the US has invested enormous amounts of money in the past decades for supercomputing capability-building; Canada is not even close, even per-capita or per-GDP.

Supercomputing in Canada

Canada has supercomputing `Centres of Excellence’ in seven regional CFI-funded Compute Canada consortia, which are also supported by and have presences in every province of Canada. These locations combine cutting-edge computing hardware with world- leading highly qualified personnel, and they already work with academic researchers in all disciplines, from aerospace engineering to astrophysics, and from waterway conservation to zoology. The consortia continue to work to build connections with industry, and this conference is one of the ways that practitioners from industry and academia can meet to exchange ideas and methods.

SciNet

SciNet is the largest supercomputing centre in Canada.  It is one of seven regional consortia of universities and colleges across Canada with mandates to provide High-Performance Computing (HPC) resources to both their own academic researchers as well as other users across the country and international collaborations. SciNet was funded by the National Platform Fund (NPF) of the Canada Foundation for Innovation (CFI) and by the Province of Ontario as well as the University of Toronto Faculties of Arts and Science, Engineering and Medicine.

SciNet represents a large and diverse group of internationally and nationally recognized researchers in a wide variety of fields of research endeavour based at the University of Toronto and the affiliated research hospitals and institutes:

Baycrest Centre for Geriatric Care Hospital for Sick Children University Health Network
Mount Sinai Hospital Ontario Institute for Cancer Research Women’s College Hospital
St. Michael’s Hospital Toronto Rehabilitation Institute Bloorview Kids Rehab
Sunnybrook Health Sciences Centre Centre for Addiction and Mental Health

All of these institutions have been involved at the forefront of high-performance computing in a vast array of scientific disciplines for decades.  The SciNet consortium, formally instated in 2007, takes full advantage of this broad experience and knowledge.

The SciNet offices are at the St. George street campus at 256 McCaul Street.  To accommodate the large floor space and power needs, the SciNet datacentre facility is housed in a warehouse about 30 km north of campus in Vaughan.   The datacentre is one of the most environmentally friendly in Canada, using a fraction of the power for  cooling of most similar facilities.