2016 Open Call for Participation in the Beacon Project

The National Institute for Computational Sciences (NICS) at the University of Tennessee (UT) is pleased to announce the 2016 open call for participation in the Beacon Project, an ongoing research project funded by the National Science Foundation and UT to explore the impact of emerging computer architectures on computational science and engineering.

Beacon is a Cray CS300-AC Cluster Supercomputer that ranked #1 on the November 2012 Green500 list. The 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. Industry-leading debuggers, profilers, and performance analysis tools from Intel, RogueWave, and Allinea are available for use on the system.

Individual researchers and teams are invited to submit proposals for projects investigating the impact of the Intel Xeon Phi coprocessor(s), the large amounts of memory per CPU core, and/or the use of parallel SSD storage 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
  • Data-driven computing and workflows
  • Novel algorithms targeting the Intel® Xeon Phi™ coprocessor
  • Novel algorithms involving hierarchical SSD storage
  • Programming languages and tools
  • Debugging and profiling tools
  • Performance evaluation studies and tools
  • Energy-efficient computing and energy-aware algorithms

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. While Beacon is primarily a development cluster, projects involving production runs at small scale are welcome.

Proposals should include the following information, using the provided Microsoft Word template as guidance:

  • 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
  • A short biography for the principal investigator, any co-investigators, and any other senior personnel involved in the proposed research
  • A listing of any references cited in the proposal

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. Proposals should not be longer than 10 pages, excluding biographies and references. Each proposal will be peer-reviewed by computational scientists upon receipt, and any associated award will be made immediately following peer-review. Proposals are still being accepted.

Selected proposals will be awarded time, training, and limited support on Beacon with an ending date of June 30, 2016. In return, award recipients are expected to disseminate their results to the scientific and engineering community in the form of papers, presentations, and a final report. Award recipients 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.