Title :
Configurable Flow Models for FPGA Particle Graphics Engines
Author :
Wong, Andrew J. ; Gross, Warren J.
Author_Institution :
Dept. of Electr. & Comput. Eng., McGill Univ., Montreal, QC, Canada
Abstract :
We describe the implementation of a hardware-accelerated particle graphics engine on a reconfigurable computer. The engine incorporates a configurable flow model that enables the simulation of complex spatially-dependent particle graphics effects. The FPGA particle engine was designed using the Mitrion-C high-level language, and did not require detailed hardware design. The engine was implemented on a SGI Altix 350 with four Xilinx Virtex-4 LX200 FPGAs. The engine achieves speedups of 35 to 58 times over a 1.5 GHz Itanium 2 CPU when using one and four FPGAs respectively.
Keywords :
computational fluid dynamics; computer graphics; field programmable gate arrays; flow simulation; FPGA particle graphics engines; Mitrion-C high-level language; Xilinx Virtex-4; configurable flow models; hardware-accelerated particle graphics engine; reconfigurable computer; Acceleration; Computational modeling; Computer graphics; Engines; Field programmable gate arrays; Hardware; High level languages; Pipelines; Table lookup; Viscosity; FPGA; Particle Graphics;
Conference_Titel :
Field-Programmable Custom Computing Machines, 2008. FCCM '08. 16th International Symposium on
Conference_Location :
Palo Alto, CA
Print_ISBN :
978-0-7695-3307-0
DOI :
10.1109/FCCM.2008.20