Indiana University

Skip to:

  1. Search
  2. Breadcrumb Navigation
  3. Content
  4. Browse by Topic
  5. Services & Resources
  6. Additional Resources
  7. Multimedia News

Media Contacts

Daphne Siefert-Herron
Information Manager, Pervasive Technology Labs at Indiana University
dsiefert@indiana.edu
812-856-1242

Last modified: Monday, November 24, 2008

IU and Technische Universität Dresden undergraduates win supercomputing competition

FOR IMMEDIATE RELEASE

November 24, 2008

BLOOMINGTON, Ind. -- Not even an ocean could stand between them and success. The "ClusterMeisters," a team of students from the Indiana University School of Informatics and Technische Universität Dresden (TUD) were awarded first place in the SC08 Cluster Challenge, an international collegiate competition for leading-edge, energy-efficient high performance computing.

ClusterMeisters

ClusterMeisters: upper row left to right: Torsten Höfler (Adviser IU), Guido Juckeland (Advisor TUD), ValkyrieSavage (IU), Chris Beckley (IU); lower row left to right: Timo Schneider (IU), Jens Domke (TUD), Jupp Müller (TUD), Robin Geyer (TUD)

Print-Quality Photo

The award was presented Thursday, November 20, at SC08, the world's largest international conference for high performance computing, networking, storage and analysis, held in Austin, Texas.

The annual Cluster Challenge competition allows undergraduate students to hone their supercomputing skills and engage in friendly competition with the best and brightest of their peers, while showcasing the ability of clusters and open-source software to solve interesting and important problems.

The team was supervised in the US by IU Computer Science Professor Andrew Lumsdaine and Torsten Hoefler of the IU Pervasive Technology Institute's Open Systems Lab, and in Germany by Professor Wolfgang E. Nagel and Guido Juckeland of TUD's Center for Information Services and High Performance Computing.

A variety of challenge events are staged annually in conjunction with the SC conferences. This year's Cluster Challenge win marks the fifth time IU has won an SC challenge, while the TUD team has claimed a total of four. Of these wins, three have included collaborations between the two schools.

To prepare for this year's Cluster Challenge, the US and German students used videoconferencing technology to collaborate for several weeks before finally meeting face to face in Austin.

"Working with our partners from IU posed quite a challenge because the time to work together on a problem is very limited; it meant lots of late nights," said team captain Jupp Müller, a senior in Information System Technology at TUD. "But our effort really paid off; it felt like all the pieces fell into place during the competition."

Once in Austin, the team worked in two groups that alternated 12 hour shifts for the four-day day duration of the challenge.

"Despite having an ocean separating the two halves of our team, the quality of our team members and the communication infrastructure we set up allowed for strong working relationships and led to great teamwork and friendships - especially once we all met in Austin," said team member Chris Beckley, an IU junior studying computer science. "Even if we had ended up last place, I would be just as proud to say I was part of such a wonderful, hard-working team."

Working with corporate partners IBM and Myricom, the student team developed the most powerful system in the competition while not exceeding the maximum electrical power allowed under challenge rules. IBM provided a new System X iDataplex server, designed to use up to 40% less power than similarly configured standard servers. This provided the team with a cluster that has a peak performance of over 1 teraFLOP. The latest Myrinet 10G equipment with 16 QSFP ports that was provided by Myricom, uses significantly less power and delivers higher application performance than other high-performance networking technologies.

"We are happy to partner with the students at Indiana University and Technische Universität Dresden in Germany in this year's Cluster Challenge," said Sergio Amoni, Director IBM System x Marketing. "Our first place finish is attributable to the extraordinary skills of the students coupled with IBM's newly introduced Intel x86-based iDataPlex dx360 server - optimized for HPC environments."

Also important to the ClusterMeisters' victory was application performance tuning, using state-of-the-art commercial compilers and advanced tools like Vampir, to analyze and improve the application behavior.

The ClusterMeisters achieved best overall performance for the High Performance Computing Challenge benchmark and five different scientific applications:

  • The OpenFOAM (Open Field Operation and Manipulation) CFD Toolbox can simulate anything from complex fluid flows involving chemical reactions, turbulence and heat transfer, to solid dynamics, electromagnetics and the pricing of financial options. OpenFOAM, produced by OpenCFD Ltd, is freely available and open source, licensed under the GNU General Public License.
  • The wave propagation program (WPP) developed by Lawrence Livermore National Laboratory implements substantial capabilities for 3-D seismic modeling. It is a parallel code for simulating time-dependent elastic and viscoelastic wave propagation, with some provisions for acoustic wave propagation.
  • POY4 is a flexible, multi-platform program for phylogenetic analysis of molecular and other data. An essential feature of POY is that it implements the concept of dynamic homology allowing optimization of unaligned sequences. POY4 can analyze entire chromosomes and genomes, taking into account large-scale genomic events (translocations, inversions, and duplications).
  • RAxML (Randomized Axelerated Maximum Likelihood) is a program for sequential and parallel Maximum Likelihood based inference of large phylogenetic trees. It was originally been derived from fastDNAml which, in turn, was derived from Joe Felsentein's dnaml, part of the Phylogeny Inference Package (PHYLIP).
  • The General Atomic and Molecular Electronic Structure System (GAMESS) is a general ab initio quantum chemistry package. It can be used to compute the properties of molecules and chemical reactions using a wide variety of theoretical models.