Hostname: page-component-7b7567c8c-tnkh4 Total loading time: 0 Render date: 2026-11-16T13:44:15.985Z Has data issue: false hasContentIssue false

Current Status of GRAPE Project

Published online by Cambridge University Press:  01 September 2007

J. Makino*
Affiliation:
Division of Theoretical Astronomy, National Astronomical Observatory of Japan, 2-21-1 Osawa, Mitaka, Tokyo, 181-8588 email: makino@cfca.jp
Rights & Permissions [Opens in a new window]

Abstract

Core share and HTML view are not available for this content. However, as you have access to this content, a full PDF is available via the 'Save PDF' action button.

I'll summarize the current status of GRAPE project. GRAPE-6, completed in 2002, has been used by a number of people, for a wide variety of problems such as planet formation, star cluster dynamics, galactic nuclei, and cosmology. In 2004, we started the development of the next-generation machine, GRAPE-DR. GRAPE-DR has a architecture radically different from that of previous GRAPEs. It does not have hardwired pipeline for gravitational force calculation but a large number of small and simple programmable processors. This change made it possible to apply GRAPE-DR to a wide range of problems to which GRAPE was not efficient, and at the same time it helps us to explore new algorithms for N-body simulations. The GRAPE-DR chip was completed in 2006, and second prototype board was completed in May 2007. We hope to have full production-level board commercially available by the end of year 2007. A single board will offer the theoretical peak speed of 2 Tflops, about 20 times as that of a single PCI card version of GRAPE-6.