Performance and accuracy of a GRAPE-3 system for collisionless N-body simulations

Details Performance and accuracy of a GRAPE-3 system for collisionless N-body simulations Performance and accuracy of a GRAPE-3 system for collisionless N-body simulations

1997-09-25

arxiv.org/abs/astro-ph/9709246v1

Astronomy and Astrophysics

Astrophysics

13 pages Latex, with 11 figures, accepted for publication in MNRAS

Scientific paper

10.1046/j.1365-8711.1998.01102.x

The performance and accuracy of a GRAPE-3 system for collisionless N-body simulations is discussed. After a description of the hardware configurations available to us at Marseille, and the usefulness of on-line analysis, we concentrate on the performance and accuracy of the direct summation and tree code software. For the former we discuss the sources of round-off errors. The standard Barnes-Hut tree code has been modified for GRAPE-3, by dividing particles into blocks and performing the tree traversal for the whole block. The forces are then calculated by direct summation over the whole interaction list. The performance of the tree code depends on the number of particles in the block, the optimum number depending on the speed of the front end and the number of boards. The time per step scales with the number of particles, and decreases only weakly as the tolerance increases. We discuss the effect of the front end and compare performances with those of general purpose computers. The accuracy of both direct summation and the tree code is discussed as function of the number of particles and the softening. The increased role of direct summation in the force calculation enhances the accuracy compared to the standard tree code. Finally, we follow the evolution of an isolated barred galaxy using different hardware and software in order to assess the reliability of our results. We find excellent agreement between the pattern speed of the bar in direct summation simulations run on the high precision GRAPE-4 machines and on our GRAPE-3 system. The agreement with the tree code is very good for tolerance values smaller than about 1.0. We conclude that GRAPE-3 systems are well suited for collisionless simulations, in particular those of galaxies.

Affiliated with

Also associated with

No associations

Rating

Performance and accuracy of a GRAPE-3 system for collisionless N-body simulations does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Performance and accuracy of a GRAPE-3 system for collisionless N-body simulations, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Performance and accuracy of a GRAPE-3 system for collisionless N-body simulations will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-201388