Presentation 2008-05-22
An Implementation of High Precision Floating-point Operation Units on FPGA
Naohito NAKASATO, Tadashi ISHIKAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Kinds of numerical application programs require high-precision (better than double-precision) floating-point (FP) operations. An example of such applications is a numerical evaluation of Feynmann loop integrals. To evaluate those integrals with a desired accuracy on conventional CPUs, one utilize emulation techniques using double-precision operations or 64-bit integer operations. However, a perfomance penalty of emulation techniques is as large as 4-5% of DP operations. To tackle those problems, we are currently working on implementation of quadruple-precision floating-point operation units on a FPGA device. We report a current status of our project and propose a computation unit that is capable of high-precision numerical integration.
Keyword(in Japanese) (See Japanese page)
Keyword(in English)
Paper # RECONF2008-2
Date of Issue

Conference Information
Committee RECONF
Conference Date 2008/5/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Reconfigurable Systems (RECONF)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An Implementation of High Precision Floating-point Operation Units on FPGA
Sub Title (in English)
Keyword(1)
1st Author's Name Naohito NAKASATO
1st Author's Affiliation University of Aizu()
2nd Author's Name Tadashi ISHIKAWA
2nd Author's Affiliation High Energy Accelerator Research Organization, KEK
Date 2008-05-22
Paper # RECONF2008-2
Volume (vol) vol.108
Number (no) 48
Page pp.pp.-
#Pages 5
Date of Issue