Canada’s most powerful research supercomputer, unveiled earlier this year at U of T, will enable researchers to focus immense processing capability – equivalent to some 60,000 desktop machines – on a single, complex question. And its first use could lead to a better understanding of how the world’s oceans affect climate change.
The $18-million system – dubbed “Niagara” – is the first major upgrade at SciNet, U of T’s high-performance computing division, in a decade. With 10 times the power of its predecessor, Niagara is supported by 12 petabytes (12 million gigabytes) of storage. It’s believed to rank among the top 50 supercomputers in the world.
The system is housed in a secure, nondescript location in Vaughan, Ontario, and requires less power than its predecessor, with savings roughly equivalent to the energy used by 300 homes in a year.
For Niagara’s first test, Richard Peltier, a U of T University Professor of physics, and colleagues from the U.S. and Canada, ran a calculation related to a research question about the behaviour of oceans. It generated millions of gigabytes of data. The researchers then used the supercomputer to compare the data with observations streamed from two sensors located on the floor of the Pacific Ocean.
The results of this research will improve our understanding of ocean internal waves, which occur far below the water’s surface. Peltier hopes a deeper knowledge of these waves will improve our ability to explain the evolution of ocean temperature, salinity, circulation and marine biology – and to create better models of climate change.
Peltier, who is also the scientific director of SciNet, came up with the idea of running a “heroic calculation” on Niagara after discussing with colleagues how best to strenuously test the power of the new system.
“Devoting the entire machine, not only a portion of it, to this one calculation is what makes the calculation ‘heroic,’” he says. “This is pure, curiosity-driven research.”
By bringing artificial intelligence into chemistry, Prof. Aspuru-Guzik aims to vastly shrink the time it takes to develop new drugs – and almost everything else