Thursday, May 17 – 2:30 p.m.
One-hundred million dollars worth of new supercomputing systems are currently being deployed across the country in one of the largest renewals ever for Canadian researchers. Procuring, installing, and operating such systems for shared use by large numbers of researchers comes with a host of significant and unique challenges. Researcher requirements and associated software tools and training can change rapidly (e.g. significant adoption of machine and deep learning techniques by physicists and others). New technologies can lead to centers being on the “bleeding-edge” and large-scale deployments often encounter hardware and software problems that don’t manifest on smaller systems. Even datacenter design and infrastructure are challenged by power and cooling demands and the need to minimize operating costs.
This panel gathers together the technical experts and leaders who have been responsible for deploying and supporting Canada’s largest supercomputer systems over the past decade in order to learn from their vast experience, discuss challenges and share “war stories”.
Moderator: Chris Loken
Chief Technical Officer, Compute Ontario
Dr. Chris Loken has more than two decades of experience in technology and specifically, within high performance computing. For the past 16 years, Dr. Loken has held progressively senior roles at the University of Toronto, most recently as the Chief Technical Officer at SciNet. Prior to this role, Dr. Loken held visiting academic and research roles at New Mexico State University, University of Missouri and St. Mary’s University, Halifax. He obtained his PhD from Queen’s University in Physics and Astronomy.
Dr. Daniel Gruner
Chief Technology Officer, SciNet
Panelist Danny Gruner, Chief Technology Officer for SciNethas more than twenty years’ experience working with a variety of programming languages, parallel computing, scientific modeling, software architecture, and administration and architecture of large Beowulf clusters and large shared-memory parallel computers. He has a doctorate in chemical physics from the University of Toronto.
Chief Technology Officer, ACENET
Greg has been involved with HPC, either as a user, systems administrator, or managing executive for over 15 years. Greg was instrumental in the creation of ACENET, and as Chief Technology Officer since 2007, has guided its technical team and computing resources.
He sits on the board of ACORN-NS (Nova Scotia’s advanced network infrastructure organization) and has served as Vice-Chair on Regional Economic Development boards, both locally and provincially, with a special interest in the economic development potential of technology and smart-community initiatives.
Director of Technology, SHARCNET
Panelist John Morton is Director of Technology at SHARCNET and has a broad range of skills, including 12 years managing all aspects of the operations at SHARCNET as well as supervising a large distributed team of technical staff. Mr. Morton’s colleagues are responsible for the operations of SHARCNET’s large high-performance compute facilities, storage resources and its private wide area network responsible for creating a virtual organization that supports over 18 Colleges and Universities. Mr. Morton has a Master of Science degree in Computing and Information Science along with a Bachelor of Math in Computer Science and Applied Math.
Site Lead for the National Compute Canada system "Cedar (GP2)"
Dr. Martin Siegert is the Director, Research Computing, at Simon Fraser University and the Site Lead for Compute Canada’s Cedar system. He has more than 20 years of experience with advanced research computing and manages a growing team of experts at SFU who support the WestGrid and Compute Canada computing facilities at the university. His background is in Theoretical and Computational Condensed Matter Physics and he has a PhD in Physics from the University Hannover, Germany.
Chief Technology Officer, Calcul Québec
Manager, HPC Integration in Supercomputing at Shared Services Canada