By Topic

Grid Environment for Computational Astrophysics Driven by GRAPE-6 with HMCS-G and OmniRPC

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

5 Author(s)
T. Boku ; Graduate Sch. of Syst. & Inf. Eng., Tsukuba Univ., Ibaraki, Japan ; K. Onuma ; M. Sato ; Y. Nakajima
more authors

In computational astrophysics, the effect of gravity is essential even with very heavy computation of O(N^2 ) for N particles. Several special purpose machines have been implemented as gravity engine to handle this problem extremely fast, however there are few sites in the world to operate such systems. We have developed a grid environment to access such a system based on grid-RPC, named HMCS-G. Using HMCS-G, a multi-physical computational astrophysics simulation can be performed with the combination of a PC-cluster and the gravity engine GRAPE-6. In typical size of problems for galaxy formation, however, the computation power of PC-cluster is much weaker than that of GRAPE-6, and we need multiple sets of PC-clusters to share the power of GRAPE-6 on grid environment. We have developed such a system using OmniRPC, a grid-enabled RPC system. In this system, OmniRPC are used both to distribute jobs for parameter search on multiple PC-clusters and to access GRAPE-6 server from these clusters. We performed an actual problem to search several formations of galaxy on this system, and confirmed such a solution is useful for wide variety of computational astrophysics research.

Published in:

19th IEEE International Parallel and Distributed Processing Symposium

Date of Conference:

04-08 April 2005