The CP-PACS Project
Y. Iwasaki
Abstract
The CP-PACS project is a five year plan, which formally started in April 1992 and has been completed in March 1997, to develop a massively parallel computer for carrying out research in computational physics with primary emphasis on lattice QCD. The initial version of the CP-PACS computer with a theoretical peak speed of 307 GFLOPS with 1024 processors was completed in March 1996. The final version with a peak speed of 614 GFLOPS with 2048 processors was completed in September 1996, and has been in full operation since October 1996. We describe the architecture, the final specification, the hardware implementation, and the software of the CP-PACS computer. The CP-PACS has been used for hadron spectroscopy production runs since July 1996. The performance for lattice QCD applications and the LINPACK benchmark are given.
Create a lesson
Related papers
Confining strings in the reconfined phase of trace deformed three dimensional SU(2) Yang-Mills theory
Claudio Bonati, Michele Caselle, Alessio Negro et al.
Generalizable Equivariant Diffusion Models for Non-Abelian Lattice Gauge Theory: Analytical Foundations and Numerical Scaling Tests
Gert Aarts, Diaa E. Habibi, Andreas Ipp et al.
2+1 flavor light hadron spectrum with PACS10 configurations
K. -I. Ishikawa, N. Ishizuka, Y. Kuramashi et al.
Precision determination of Bc and Bc* meson decay constants from lattice QCD
Navdeep Singh Dhindsa, Debsubhra Chakraborty, Nilmani Mathur
Meson spectroscopy from lattice QCD
Andrew W. Jackura
Generative sampling with physics-informed kernels
Friederike Ihssen, Renzo Kapust, Jan M. Pawlowski