Open Call for Participation in the Beacon Project

The National Institute for Computational Sciences (NICS) at the University of Tennessee is pleased to announce an open call for participation in the Beacon Project, an ongoing research project partially funded by the National Science Foundation that explores the impact of emerging computer architectures on computational science and engineering.

Project teams associated with the Beacon Project are currently exploring the impact of the Intel® Xeon Phi™ coprocessor on scientific codes and libraries such as H3D, PSC, OMEN, ENZO, MADNESS, NWChem, Amber, MILC, and MAGMA through porting and optimization work on a small, experimental cluster equipped with pre-production coprocessors. To expand this effort and to facilitate research into energy-efficient supercomputing, NICS is replacing the experimental cluster with Beacon, a Cray Xtreme-X Supercomputer that ranks #1 on the November 2012 Green500 list.

The Beacon system offers access to 48 compute nodes and 6 I/O nodes joined by FDR InfiniBand interconnect providing 56 Gb/s of bi-directional bandwidth. Each compute node is equipped with 2 Intel® Xeon® E5-2670 processors, 4 Intel® Xeon Phi™ coprocessors 5110P, 256 GB of RAM, and 960 GB of SSD storage. Each I/O node provides access to an additional 4.8 TB of SSD storage. Thus, Beacon provides 768 conventional cores and 11,520 accelerator cores that provide over 210 TFLOP/s of combined computational performance, 12 TB of system memory, 1.5 TB of coprocessor memory, and over 73 TB of SSD storage, in aggregate.

Individual researchers and teams are invited to submit proposals for projects investigating the impact of the Intel® Xeon Phi™ coprocessor on areas of interest to their particular field of study. Some possible areas of interest include (but are not limited to):

  • Computational modeling and simulation
  • Data analysis and visualization
  • Novel algorithms targeting the Intel® Xeon Phi™ coprocessor
  • Programming languages and tools
  • Debugging and profiling tools
  • Performance evaluation studies and tools
  • Energy-efficient computing

Projects involving high-performance computing (HPC) applications in fields such as biology, economics, social sciences, and other non-traditional HPC domains are especially encouraged, as are projects associated with data-intensive computing and data-driven workflows.

Proposals should include the following information in 8 pages or fewer, using an easily readable 10-point or 11-point font (excluding captions, which may be smaller):

  • A brief overview that establishes the scientific context of the proposed research
  • A clear description of the proposed research and its expected impact, including clearly defined metrics for measuring the success of the effort
  • A justified estimate of the computational resources required for the project
  • A clear description of both the computational readiness of any codes or algorithms associated with the proposed research and the readiness of the research team to begin work on the project
  • A description of any outreach or educational activities associated with the proposed work

Proposals should also include the following information (subject to the noted limits):

  • A short biography for the principal investigator, any co-investigators, and any other senior personnel involved in the proposed research (2 pages per person or fewer)
  • A listing of any references associated with the proposal (4 pages or fewer)

Each proposal should be submitted as a single PDF file to beacon-proposal@nics.tennessee.edu to be considered for an award at the discretion of the project director.

Selected proposals will be awarded time, training, and limited support on Beacon for a period of one year from the award date, which is anticipated to be March 1, 2013 or shortly thereafter. In return, partners are expected to disseminate their results to the scientific and engineering community in the form of papers, presentations, and a final report. Partners are also expected to provide status updates to NICS when requested.

This material is based upon work supported by the National Science Foundation under Grant Number 1137907. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.