Today SC17 announced that 16 teams will take part in the Student Cluster Competition. Hailing from across the U.S., as well as Asia and Europe, the student teams will race to build HPC clusters and run a full suite of applications in the space of just a few days.
The Student Cluster Competition is a high energy event featuring young supercomputing talent from around the world competing to build and operate powerful cluster computers. Created as an opportunity to showcase student expertise in a friendly yet spirited competition, the SCC aims to introduce the next generation of students to the high-performance computing community. The number of teams selected for SC17 is the highest since the first event was held at SC07 in Reno, and the 19 reviewers sifted through 20 team applications, also the highest number to date.
“We want to expand the reach of the competition,” said SC17 SCC Chair Stephen Harrell of Purdue University. “In order to improve the competition we need to make it more competitive, as well as more available to a broader range of schools and teams. One of the ways we are working to do that is by bringing on more teams.”
Among the SC17 teams are three that have never competed before: a team from Poland comprising the University of Warsaw, Lodz University of Technology and the Warsaw University of Technology; the Georgia Institute of Technology (Georgia Tech); and William Henry Harrison High School of Indiana. Two other high schools, Skyline High in Utah and Maine South High School will also be competing for the first time, partnering with nearby universities. Southwestern Oklahoma State University will also make its first foray into the competition, partnering with time-tested San Diego State University.
The SC17 competition will also mark the first time that the winners of the world’s top three student cluster competitions will meet: SC16 winner University of Science and Technology of China; Tsinghua University of China, winner of the ASC17 Student Supercomputer Challenge; and the winner of the upcoming ISC Student Cluster Competition to be held in June.
2017 Student Cluster Competition Teams:
- Friedrich-Alexander University/Technical University of Munich, Germany; in their fourth year of competition,
- Georgia Institute of Technology, first year,
- Illinois Institute of Technology/Maine South High School, third time for Illinois Tech,
- MGHPCC Green Team (UMass Boston/UMass Lowell/Harvard University), fourth time,
- Nanyang Technological University, Singapore, second time
- National Tsing Hua University, Taiwan, ninth time for the team that won top honors at SC10 and SC11,
- Northeastern University, Massachusetts, third time,
- Peking University, China, second time
- San Diego State University/Southwestern Oklahoma State University, second time for San Diego State,
- Tsinghua University, China (The winning team from the ASC17 Student Supercomputer Challenge), second time,
- University of Science and Technology of China (The winning team from the SC16 Student Cluster Competition), third time,
- University of Texas at Austin/Texas State University, fourth time for UT Austin, second time for this pairing,
- University of Utah/Skyline High School, second time for Utah,
- University of Warsaw/Lodz University of Technology/Warsaw University of Technology, Poland, first time for a team from Poland,
- William Henry Harrison High School, Indiana, first time,
A new wrinkle introduced at SC16 is the SCC Reproducibility Initiative. This component of the competition challenges students to reproduce a paper rather than run prescribed data sets. For this year, the SCC committee chose “The vectorization of the Tersoff multi-body potential: An exercise in performance portability” by Markus Höhnerbach, Ahmed E. Ismail and Paolo Bientinesi for the 2017 reproducibility challenge. One unique aspect of this paper is that it reports on optimizations for several hardware architectures.
We want students to understand, early in their careers, the important role reproducibility plays in research,” said SC17 SCC Chair Stephen Harrell of Purdue University. “Reproducibility is a big topic in both domain science and HPC and we want students to learn about both aspects of this topic.”