CHAMPAIGN, IL —
The National Computational Science Alliance (Alliance) began offering a new web-based
course on the Message Passing Interface (MPI) in mid-September.
The new course was developed by the Alliance's Partners for Advanced Computational
Services (PACS) based on the results of Alliance user surveys conducted in 1999 and
2000. In both surveys Alliance users said they were interested in receiving web-based
training on MPI because web-based training is self-paced, convenient, and available
at no cost to anyone with Internet access. Staff at five PACS sites worked together
to develop the training program: the National Center for Supercomputing Applications
(NCSA) at the University of Illinois, University of New Mexico, Boston University,
University of Kentucky, and the Ohio Supercomputing Center (OSC). Leslie Southern
at OSC led the development effort.
MPI is a standard library of Fortran subroutines, or C language function calls,
that implement a message-passing program, thereby making distributed computing
possible. With MPI, a programmer can write code that will work on any computing
system on which the MPI library is installed without having to make changes. MPI
programs can be used and compiled on a wide variety of parallel computers with
either distributed or shared memory, such as the IBM SP2, the SGI Origin2000, or
a cluster of workstations.
The new MPI course uses webCT, a web-based course tool originally developed by the
University of British Columbia. The webCT environment provides a variety of online
learning tools, including help, search tools, a course discussion space, self-tests,
and a glossary of terms. The course was extensively tested over the summer, and
additional materials were incorporated based on feedback from test groups.
Each of the 13 sections in the MPI course offers a statement of objectives and a
brief introduction. Most sections include examples and descriptions of routines
as well as the most appropriate applications or algorithms. One case study is
used throughout the course, building on the new materials introduced in each
section. Self-tests in each section help users check their understanding of the
material presented.
Southern noted that this asynchronous high-performance training builds on the
Alliance's concept of a Virtual Machine Room (VMR). The Alliance's VMR project
is an effort to bring together the computational and data resources of the Alliance
no matter what their physical locations. These resources are made accessible through
a common interface using interoperable component technologies and tools. The user
sees a single coherent system of Alliance resources that appear to be managed by,
and located at, one site.
"Our asynchronous training lets researchers control their training environment from
any place, at any time, and at their own pace," Southern said. "PACS is pleased to
answer the call from users expressed in the recent user survey for such training and to
provide our first course on the most commonly listed topic—MPI."
Users can register for the course through the WebCT-HPC
website. Once registered, a user keeps the same login ID and password to take other
courses available on the site. All course materials are available from the home page.
The National Computational Science Alliance is a partnership to prototype an
advanced computational infrastructure for the 21st century and includes more than
50 academic, government and industry research partners from across the United States.
The Alliance is one of two partnerships funded by the National Science Foundation's
Partnerships for Advanced Computational Infrastructure (PACI) program, and receives
cost-sharing at partner institutions. NSF also supports the National Partnership for
Advanced Computational Infrastructure (NPACI), led by the San Diego Supercomputer Center.
The National Center for Supercomputing Applications is the leading-edge site for the
National Computational Science Alliance. NCSA is a leader in the development and
deployment of cutting-edge high-performance computing, networking, and information
technologies. The National Science Foundation, the state of Illinois, the University of
Illinois, industrial partners, and other federal agencies fund NCSA.