Astronomisches Rechen-Institut Heidelberg

ARIPRINT:    Database of publications of the institute

Preprint No. 94

Author(s): Kuberka, T., Kugel, A., Maenner, R., Singpiel, H., Spurzem, R., Klessen, R.
Title: AHA-GRAPE: Adaptive Hydrodynamic Architecture - GRAvity PipE
Source: to appear in Proc. of The 1999 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA 99), Las Vegas, USA
Year: 1999
Abstract: In astrophysics numerical star cluster simulations and hydrodynamical methods like SPH require computational performance in the petaflop range. The GRAPE family of ASIC-based accelerators improves the cost-performance ratio compared to general purpose parallel computers, however with limited flexibility. The AHA-GRAPE architecture adds a reconfigurable FPGA-processor to accelerate the SPH computation. The basic equations of the algorithm consist of three parts each scaling with the order of O(N), O(N*Nn), and O(N2) respectively, where N is in the range of 104 to 107 and Nn approximately 50. These equations can profitably be distributed across a host workstation, an FPGA processor, and a GRAPE subsystem. With the new ATLANTIS FPGA processor we expect a scalable SPH performance of 1.5 Gflops per board. The first prototype AHA-GRAPE system will be available in mid-2000. This 3-layered system will deliver an increase in performance by a factor of 10 as compared to a pure GRAPE solution.
Preprint issued: May 1999
Remarks:

)Full text of the preprint in the ARIPRINT in the following format:
[Postscript]


Back to Preprint Series (overview) or Publications or Homepage


Letzte Änderung/Updated: 02.08.2004

ariprint@ari.uni-heidelberg.de