| Co-Investigator(Kenkyū-buntansha) |
デシック ビクター UCLA, 理学部, 教授
ドーソン ジョン UCLA, 理学部, 教授
NISHIKAWA Kenichi Louisiana State Univ.Dept.of Physics and Astronomy Assitant Professor, 理学部, 助教授
ITO Toshiaki Univ.of Tokushima, Faculty of Integrated Arts and Sciences Assistant Professor, 総合科学学部, 助教授 (60201927)
IKEBE Yasuhiko Univ.of Aizu, School of Computer Science and Engineering Professor, コンピュータ理工学部, 教授 (10114034)
DAWSON John UCLA,Dept.of Physics Prefessor
DECYK Victor K. UCLA,Dept.of Physics Prefessor
|
|---|
| Research Abstract |
Particle-in-Cell (PIC) or Particle-in-Mesh (PIM) code has been parallelized in many advanced parallel computers. On the other hand, many new high-performance PCs as those based on Pentium II or Alpha 21264 RISC processors have been introduced recently. In a due course, these high-performance PCs have a high potential for the high-performance-computing purposes. We are currently building a dual PentiumPro PC cluster for a space weather simulator. In this report, using a skeleton-PIG-code that is developed by Professor V.K.Decyk of UCLA for a testbed and benchmarking purposes, we have measured the performances of the Skelton PIC code using PVM, pgHPF, and OpenMP on these personal PC, after optimizing the code for PentiumPro. A skeleton-PLC-code has been proposed by Decyk as a testbed where new algorithms can be developed and tested, and new computer architecture can be evaluated. This code has been deliberately keptminimum, but they include all the essential pieces for which algorithms n
… More
eed to be developed.
The code contains the critical pieces needed for depositing charge, advancing particles, and solving thefield. The code moves only electrons, with periodic electrostatic forces obtained by solving Poisson's equation with the fast Fourier transforms. The code uses the electrostatic approximation, and magnetic fields are neglected. The only diagnostic is particle and field energy. In the present report, we build a PC cluster using 16 HP Vectras 61200 that are cheap PCs based (about \$ 3,500 each) on dual PentiumPro, and interconnect them through a cheap HP 10BaseT ethernet switch. We optimize the skeleton-PIG-code for PentiumPro using a common RISC optimization method. We benchmark the skeleton-PLC-code on the PC cluster and compare the performance with other commercial parallel computers. Since our PC cluster is the distributed shared memory system i.e.dual PentiumPro cluster, we use PVM message passing between PCs and use OpenMP within a PC for the communication for sending particle and field data in skeleton-PlC-code. Less
|
