Early Evaluation of the Cray X1
Proceedings of the 2003 ACM/IEEE conference on Supercomputing
Evaluation of Cache-based Superscalar and Cacheless Vector Architectures for Scientific Computations
Proceedings of the 2003 ACM/IEEE conference on Supercomputing
A performance evaluation of the cray x1 for scientific applications
VECPAR'04 Proceedings of the 6th international conference on High Performance Computing for Computational Science
Investigation of leading HPC I/O performance using a scientific-application derived benchmark
Proceedings of the 2007 ACM/IEEE conference on Supercomputing
Throttling I/O streams to accelerate file-IO performance
HPCC'07 Proceedings of the Third international conference on High Performance Computing and Communications
Hi-index | 0.00 |
The Cosmic Microwave Background (CMB) is a snapshot of the Universe some 400,000 years after the Big Bang The pattern of anisotropies in the CMB carries a wealth of information about the fundamental parameters of cosmology Extracting this information is an extremely computationally expensive endeavor, requiring massively parallel computers and software packages capable of exploiting them One such package is the Microwave Anisotropy Dataset Computational Analysis Package (MADCAP) which has been used to analyze data from a number of CMB experiments In this work, we compare MADCAP performance on the vector-based Earth Simulator (ES) and Cray X1 architectures and two leading superscalar systems, the IBM Power3 and Power4 Our results highlight the complex interplay between the problem size, architectural paradigm, interconnect, and vendor-supplied numerical libraries, while isolating the I/O filesystem as the key bottleneck across all the platforms.