Centre For Development of Advanced Computing
National PARAM Super Computing Facility
Parallel Computing Optimizing Performance of Parallel Programs
Centre for Development of Advanced Computing
A Scientific Society of the Ministry of Communication and Information Technology, Government of India
About The Workshop
Centre for Development of Advanced Computing (C-DAC) is India's national initiative in Advanced Computing set up with a mission to deliver state-of-the-art, open architecture, scalable high performance parallel computers in the desktop to teraflop range, embodying the emergent industry-standard building blocks. Towards fulfillment of this goal, C-DAC has advented the OpenFrame Architecture which has been realized in it's celebrated PARAM series of High Performance Computers.
C-DAC has established the "National PARAM Supercomputing Facility" (NPSF) with a view to extend the same to various industries and other institutions that need such a facility to process their diverse applications. NPSF currently houses the PARAM 10000, having a peak computing power of 100 Giga floating operations per second based on cluster architecture. It has 36 compute nodes and 4 server nodes. Each node is a quad-processor SMP based on SUN Microsystems's UltraSparc architecture, operating at 300 Mhz. PARAM 10000 supports different system area networks viz., Fast Ethernet, Myrinet, PARAMNet and High Performance Computing and Communications (HPCC) software which provides a flexible parallel and distributed software environment. The HPCC software suite of products includes high performance compilers, parallel debuggers, data visualizers, performance profilers and software engineering tools. Optimized MPI implementation and high performance communication protocols provide the right framework to extract maximum performance.
C-DAC has been conducting Parallel Computing Workshops, which have attracted participation from reputed researchers from all over India.
The objective of PCOPP-2002 is to understand important issues in the performance of parallel programs for large-scale scientific and engineering applications. PCOPP-2002 will emphasize the performance of serial and parallel programs using different parallel programming paradigms on PARAM 10000.
The workshop will include 16 classroom lecture sessions on Optimizing Performance of Parallel programs, 3 keynote speeches on emerging topics of high performance computing from an application perspective, and 4 half-day hands-on sessions. Hands-on session will be conducted on PARAM 10000 in the afternoon session on each day. Parallel programs using F90/MPI/HPF/OpenMP/Pthreads will be made available to participants in order to understand practical aspects of parallel computing from the performance point of view.
The workshop is meant for beginners as well as advanced level users of parallel computers. PCOPP-2002 provides a forum for HPC researchers, faculty members, research scholars and students to exchange ideas/results related to performance issues of parallel programs.