30th July 2013 Ottawa, Canada

Canadian High Performance Computing and Big Data Capabilities

Image courtesy of ZDNet
Image courtesy of ZDNet

National infrastructure for supercomputing-powered research
High performance computing (HPC) has become an indispensable and enabling technology across all research fields. Some fields have a long history with computing such as computer science, astrophysics, weather (climate modeling) and engineering. In Canada, researchers are managing and analyzing immensely large and complex data sets from sources such as medical and neuro-imaging (e.g. CBRAIN), the Large Hadron Collider (ATLAS), the Sudbury Neutrino Observatory, the Ontario Cancer Biomarker Network, Genome Canada, GeoChronos (Earth observations), telescope arrays (SKA and ALMA) and Digital humanities.

Canadian investment into high performance computing began in 1995 with the advent of c3.ca, a national organization founded by universities and the National Research Council to advocate high performance computing in Canada. Major funding from the Natural Science and Engineering Research Council provided technical staff support, and the creation of the Canada Foundation for Innovation in 1997 provided a mechanism to fund large-scale academic computing facilities. These initiatives led to the formation of several regional high performance computing consortia across Canada.

In 2005, the National Research Council, c3.ca, CANARIE, the Canada Foundation for Innovation and the research granting councils came together to develop a long-range plan (LRP) for HPC in Canada. The plan provided a compelling case for investment into HPC facilities, which in turn lead to two significant outcomes:

1. Creation of Compute Canada, a new national organization to oversee the network of HPC systems across Canada

2. Launch of the National Platforms Funds competition in 2006 by the Canadian Foundation for Innovation for the development of a collective national capability in HPC

Compute Canada functions both as an advocacy group and as a governing body to ensure the proposed platform functions as a national infrastructure. It integrates high performance computers, data resources and tools, and academic research facilities around the country to facilitate the use of HPC in Canadian research to advance scientific knowledge and innovation. This integrated network represents close to a petaflop of computing capability and online as well as long term storage.

Since the LRP, The Canadian Foundation for Innovation and provincial funding agencies have committed almost $240 million to HPC. The investment has resulted in the establishment of four major regional computing collaborations in Canada: Compute Atlantic (Atlantic Canada), Calcul Quebec (Quebec), Compute Ontario (Ontario), Compute West (Saskatchewan, Manitoba, Alberta, British Columbia)

Notable developments in Canadian big data initiatives
The latest investment info high performance computing and big data includes the development of the Southern Ontario Super Computing Innovation Platform in April 2012. The new Ontario-based $210 million dollar research and development initiative was launched by contributions from IBM ($175M), Ontario’s Ministry of Economic Development and Innovation ($15M), Ontario Centres of Excellence ($7M) and the Federal Economic Development Agency for Southern Ontario ($20M).

The consortium of seven universities lead by the University of Toronto and Western University, and supported by IBM technology backbone, will pair academics with IBM researchers to accelerate the commercialization of new discoveries. To date, 34 projects have been initiated in the past 12 months, with main focus in healthcare challenges, including neurological disorders; rapid urbanization and aging city infrastructure; and resources management, including water and energy systems.

The consortium will benefit from the installation of two IBM Blue Gene/Q supercomputers, designed to solve large-scale problems associated with scientific research, at the University of Toronto’s SciNet Datacentre, as well as cloud and agile computing infrastructure and software at the Western University.

The overall aim of this initiative is to increase Canadian competitiveness in the global economy. Indeed, the host centre, SciNet, is Canada’s largest supercomputing centre comprised of a consortium of the University of Toronto and affiliated hospitals. The two Blue Gene/Q systems located at SciNet top Canada on Green500, Graph500, and Top500, and are therefore Canada’s biggest, fastest, and greenest supercomputing platforms. The BlueGene/Q, Power BGC 16C 1.600GHz ranked 77th on the 2013 Top500 list and is the only Canadian HPC in top one hundred of the list.

Science & Innovation Network efforts on big data
The Science & Innovation team has been busy planning a big data session at the upcoming Canadian Science Policy Conference. The session will bring together experts from the UK (Hartree Centre, DiRAC, EPCC), Canada (IBM Canada R&D Centre, OCE) and the US (IBM USA) to discuss how big data capabilities offer solutions for the ‘grand challenges’ faced by today’s modern society. This will also be a great opportunity for the UK, which is looking to establish a national e-infrastructure and develop energy efficient software, to gather knowledge and share best practices with the Canadian and US experts.