Search engine for discovering works of Art, research articles, and books related to Art and Culture
ShareThis
Javascript must be enabled to continue!

GRAPE-6A: A Single-Card GRAPE-6 for Parallel PC-GRAPE Cluster Systems

View through CrossRef
Abstract In this paper, we describe the design and performance of GRAPE-6A, a special-purpose computer for gravitational many-body simulations. It was designed to be used with a PC cluster, in which each node has one GRAPE-6A. Such a configuration is particularly cost-effective in running parallel tree algorithms. Though the use of parallel tree algorithms was possible with the original GRAPE-6 hardware, it was not very cost-effective since a single GRAPE-6 board was still too fast and too expensive. Therefore, we designed GRAPE-6A as a single PCI card to minimize the reproduction cost and to optimize the computing speed. The peak performance is 130 Gflops for one GRAPE-6A board and 3.1 Tflops for our 24 node cluster. We describe the implementation of the tree, TreePM and individual timestep algorithms on both a single GRAPE-6A system and GRAPE-6A cluster. Using the tree algorithm on our 16-node GRAPE-6A system, we can complete a collisionless simulation with 100 million particles (8000 steps) within 10 days.
Title: GRAPE-6A: A Single-Card GRAPE-6 for Parallel PC-GRAPE Cluster Systems
Description:
Abstract In this paper, we describe the design and performance of GRAPE-6A, a special-purpose computer for gravitational many-body simulations.
It was designed to be used with a PC cluster, in which each node has one GRAPE-6A.
Such a configuration is particularly cost-effective in running parallel tree algorithms.
Though the use of parallel tree algorithms was possible with the original GRAPE-6 hardware, it was not very cost-effective since a single GRAPE-6 board was still too fast and too expensive.
Therefore, we designed GRAPE-6A as a single PCI card to minimize the reproduction cost and to optimize the computing speed.
The peak performance is 130 Gflops for one GRAPE-6A board and 3.
1 Tflops for our 24 node cluster.
We describe the implementation of the tree, TreePM and individual timestep algorithms on both a single GRAPE-6A system and GRAPE-6A cluster.
Using the tree algorithm on our 16-node GRAPE-6A system, we can complete a collisionless simulation with 100 million particles (8000 steps) within 10 days.

Related Results

British Food Journal Volume 44 Issue 11 1942
British Food Journal Volume 44 Issue 11 1942
The question whether grape juice may or may not be preserved with sulphur dioxide is one which arises occasionally as a result of a certain ambiguity in the wording of the First Sc...
Inheritance of Cluster Headache and its Possible Link to Migraine
Inheritance of Cluster Headache and its Possible Link to Migraine
SYNOPSIS We evaluated the possibility that cluster headache may be a transmitted disorder, influenced by migraine genetics. In the first part of a two part study,...
Constructing a VANET based on cluster chains
Constructing a VANET based on cluster chains
SUMMARYThe paper proposes a scheme on constructing a vehicular ad‐hoc network based on cluster chains. In the cluster construction algorithm, the distance from a potential cluster ...
Ciudad de Museos: clústeres de museos en la ciudad contemporánea
Ciudad de Museos: clústeres de museos en la ciudad contemporánea
En nuestra cultura el museo ocupa un lugar privilegiado simbólicamente, pero también físicamente, en la ciudad. Y no tan sólo lo ocupa, sino lo crea, lo define, lo cambia y le da s...
Evaluation of genetic divergence in Barley (Hordeum vulgare L.) germplasms
Evaluation of genetic divergence in Barley (Hordeum vulgare L.) germplasms
Thirty genotypes of wheat were evaluated for assessing genetic divergence among eleven different characters across one environment for exploitation in a breeding programme for impr...
Regional directions of the cluster development strategy in the field of tourism and hospitality
Regional directions of the cluster development strategy in the field of tourism and hospitality
The monograph consists of an introduction, 5 chapters, lists of used sources for each chapter separately; contains 31 tables and 37 figures. The monograph examines the theoretical ...

Back to Top