FPGA-based, floating-point reduction operations
Michael R. Bodnar, James P. Durbano, John R. Humphrey, Petersen F. Curt, Dennis W. Prather · 2006
Abstract:- Floating-point reduction operations are a vital part of scientific computational kernels, such as vector dot-products, discrete cosine transforms (DCT), and matrix-matrix multiplications. As FPGAs continue to gain popularity in custom and embedded computing platforms, implementations of these applications in such platforms are desirable. Due to the inherently deep pipelines of high-performance floating-point cores in FPGAs, reduction circuits require special feedback and buffering schemes in order to realize full throughput. In this paper, we present our floating-point reduction architecture, clocked at more than 150 MHz targeting a Xilinx Virtex2 8000-4 FPGA.