PARAM super computer review:

The Top500 project was started in 1993 to provide a reliable basis for tracking and detecting trends in high-performance computing. Twice a year, a list of the sites operating the worlds 500 most powerful supercomputer systems is assembled and released. The best performance on the Linpack benchmark is used as performance measure for ranking the computer systems. The list contains a variety of information including the system specifications and its major application areas.

The 21st Top500 List was introduced during the International Supercomputer Conference (ISC2003) in Heidelberg, June 24-27, 2003. C-DAC's Tera Scale Supercomputer PARAM Padma figures in this list at rank number 171. It is the first supercomputer to find place in the Top500 list which is indigenously designed, developed and commissioned in India.

PARAM Padma Configuration used for Top500:

PARAM Padma is a cluster of 62 4-way, IBM pSeries P630 nodes, interconnected through a high performance System Area Network, PARAMNet-II, designed and developed by C-DAC. The cluster has 496 GB of aggregate Memory and 5 TB of internal storage. PARAM Padma also has one 32 way P690 SMP node for development purpose. The PARAM Padma is powered by a suite of C-DAC's software tools.

PARAM Padma Technical Specifications

HPL Performance:

High Performance LINPACK(HPL) is a popular benchmark suite to evaluate the capabilities of Super Computers and Clusters and involves solving a system of linear equations. The results of this benchmark are published semi-annually in the list of the world's Top500 most powerful supercomputers. The Top500 test with optimal parameters on 62 node (248 processors) configuration has resulted in approximately 532 GFlops to the peak performance of 992 GFlops on PARAM Padma. The HPL performance thus is 53.6% of peak performance and the matrix size used is 224000. PARAMNet interconnect has been found to perform significantly better in all HPL tests and scales very well upto 62 nodes (248 processors) and beyond.

The description of PARAM Padma cluster used in HPL tests is as follows:

Nodes	62 nos. of IBM p630
CPU	IBM POWER4 @ 1GHz
Main Memory	496 GB
Interconnect	2.5 Gbps Full-duplex Switched PARAMNet-II network One PARAMNet NIC Card per node
OS	IBM AIX version 5.1L
System Software	Virtual Interface Provider Library (VIPL)
MPI	C-MPI (C-DAC MPI)
HPL	HPL - 1.0 - September 27, 2000 Authored by: Antoine P. Petitet University of Tennessee, Knoxville Innovative Computing Laboratories for further reading click the given link below: http://www.cdac.in/index.aspx?id=hgc