Intel Sponsors the LittleFe Project
The Intel Academic Community and Intel Labs are committed to finding creative ways to help professors integrate parallelism into their classrooms. The LittleFe project led by two of our community VIP's is one more way to introduce affordable hardware for teaching purposes. Buildout workshops in different regions and at major events give interested professors the hands-on experience they need to bring this unique cluster into their classrooms. Recent events include many of the Intel Academic Community members.
Dr. Charlie Peck
Professor Tom Murphy
Contra Costa Community College
Many institutions and teaching environments do not have access to parallel platforms for parallel and distributed computing education. Teaching key concepts such as speedup, efficiency, and load balancing are much more effectively done on a parallel platform. LittleFe is a complete 6 node Beowulf style portable computational cluster which supports shared memory parallelism (OpenMP), distributed memory parallelism (MPI), and GPGPU parallelism (CUDA).
The portable LittleFe mini-cluster is ideal for classroom demonstrations of High Performance Computing (HPC), parallel programming, and Computational/Data Enabled Science and Engineering (CDESE). Supporting shared memory, distributed memory, and GPGPU parallelism, LittleFe's secondary use is as a production HPC resource for small institutions that aren't yet able to afford or support a full scale cluster; in fact, LittleFe can be used as a gateway to, and development platform for, full scale HPC resources.
LittleFe Buildout Events
LittleFe workshops were recently offered to professors at select Colleges and Universities including several of our own Intel Academic Community members. The workshop participants got to bring home their own mini-clusters to use for teaching parallel programming, cluster computing and/or computational and data-enabled science & engineering to students and colleagues!
The buildout events for 2011 included the July Intermediate Parallel Programming and Cluster Computing weeklong workshop held at both the Norman campus of the University of Oklahoma, as well as the Polytechnic Universtiy of Puerto Rico Parallel. Sponsored by the National Computational Science Institute (NCSI, http://www.computationalscience.org/)
A second series of buildouts will take place November 13-16 at SC11 Supercomputing Conference (http://sc11.supercomputing.org) in Seattle, WA, as part of the Education Program's Intermediate Parallel Programming and Cluster Computing.
Academic Community Member Participants
- Vojislav Stojkovic, Morgan State University
- Ed Carr, North Carolina A&T
- Edin Hodzic, Santa Clara University
- Gang-Rhung, Boise State
- Vivek Sarkar, Rice University
- Luis M Vicente, Polytechnic University of Puerto Rico
- Richard Brown, St. Olaf College
- Elizabeth Shoop, Macalester College
- David Valentine, Slippery Rock University
More information about the Little Fe Project can be found at www.littlefe.net