Raspberry Jam Youth Workshop
Compute Canada, in an effort to promote and encourage the next generation of advanced research computing experts, is hosting a special event prior to the start of CANHEIT | HPCS 2016.! The Raspberry Jam Youth Workshop will introduce youth to computer programming in a fun and creative way. No prior background or experience necessary! Raspberry Jam events are a perfect introduction to the unlimited possibilities of scientific computing and programming.
What is a Raspberry Pi?
The Raspberry Pi is a low cost, credit-card sized computer that plugs into a computer monitor or TV, and uses a standard keyboard and mouse. It is a capable little device that enables people of all ages to explore computing, and to learn how to program in languages like Scratch and Python.
What happens at a Raspberry Jam event?
Teams of four will participate in a full-day workshop and create their own computer and program. The day concludes with a showcase of each team’s creation and each participate receives a special certificate.
Want more information?
Hosted by Compute Canada’s President and CEO Mark Dietrich and members of the senior management team, this Compute Canada Town Hall session will provide an open forum to discuss current Compute Canada activities and answer questions from the audience.
If you require any additional information on how to participate in the Town Hall session, please contact firstname.lastname@example.org.
When you register, please indicate your preferred method of participation:
SEEING BIG is back! Now in its second year, the SEEING BIG Data Visualization Showcase is a high-profile exhibit of state-of-the-art visualizations from Canadian researchers in any advanced research computing (ARC) field.
We’ve received top-quality submissions from researchers across Canada and will be displaying a selection of the entries on a large high-definition screen at the event venue for the duration of the conference. Following CANHEIT | HPCS 2016, submissions to SEEING BIG will be added to Compute Canada’s image and movie showcase on its website (with credits attributed to the authors).
If you have questions about the SEEING BIG Data Visualization Showcase, please contact Alex Razoumov, WestGrid Visualization Coordinator, email@example.com.
Join the University of Alberta Operations Team on a guided tour of the data centre in the General Services Building and an opportunity to have all your IT, storage and compute questions answered by our in-house experts. Meet at the Registration Desk to join the tour. Space is limited to 20 people and is first-come, first-served.
Test your luck against this poker playing computer program powered by “unbeatable” artificial intelligence software created by the Computer Poker Research Group at the University of Alberta. Heads-up Limit Texas Hold’em poker is the first imperfect information game played competitively by humans that has been solved by a computer system.
Named Cepheus, this system based at the University of Alberta is the result of over a decade of research by the Computer Poker Research Group and a joint effort with Oskari Tammelin, a Finnish software developer. The team is led by Dr. Michael Bowling, a professor of Computer Sciences at the University of Alberta, and included students Neil Burch, Michael Johanson, and Oskari Tammelin.
Cepheus teaches itself to play a game by using the CFR+ (Counterfactual Regret Minimization) algorithm. It repeatedly plays games against itself, and in the limit, converges towards a Nash Equilibrium: a perfect, unbeatable strategy that can be used against any opponent. Whenever it plays a game against itself, it looks back over the decisions it made and determines which ones were mistakes: for example, spots where it “regrets” its decision to call instead of bet, because betting would have won more money. It then changes its strategy at that spot to choose the better actions more often, in order to minimize the amount of regret it will have in its future games.
Beating the game was made possible by current and past members of the University of Alberta’s Computer Poker Research Group and with nearly a CPU millennium of computation provided by Compute Canada.
(presented by Obsidian Strategics)
One approach to scaling supercomputing performance is to harness multiple machines in data processing pipelines. In this Geo-Pipelining demonstration, a genomics data flow is streamed systolically through clusters residing in Warsaw and Poznań (Poland), Reims (France), Singapore and Edmonton with very high efficiency.
The underlying 10Gbits/s optical connectivity spans multiple research networks including Cybera and CANARIE in Canada to seamlessly aggregate each cluster’s internal InfiniBand fabrics using Obsidian’s Longbow range extenders and Crossbow routers – the InfiniCortex infrastructure. InfiniCortex is managed by Bowman Global Fabric Controllers providing a fast, segmented, scalable and secure supercomputer fabric spanning the globe. In this case, the distributed application and storage coordination is provided by middleware called InfiniCloud running on OpenStack, developed in Singapore and Australia.In the configuration reflected in this demonstration, InfiniCortex comprises five InfiniBand subnets linked with 10Gbits/s hardware encrypted paths spanning more than 28,000km.
This HPCS/CANHEIT demonstration is concurrently showcased at the International Supercomputing Conference in Frankfurt, Germany.