2009•Unpublished venueRequires access

3D finite-difference time-domain simulations of well-logging problems on graphic processing unit

Alexandra Mendoza, Rui Qiang, Ji Chen, Dagang Wu

Abstract

Three dimensional Finite-Difference Time-Domain (FDTD) code is implemented on graphics processing units (GPU) to reduce the overall simulation time. NVIDIA's Compute Unified Device Architecture (CUDA) is used for the implementation of the code on a GTX 260 graphics card. It is observed that the technique can significantly reduce the computational time for parallel FDTD simulations by a factor over 30.

About this research paper

What this paper is about

Why it matters

OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Key concepts: Computer science, Logging, Unit (ring theory), Finite difference method, Domain (mathematical analysis), Computational science, Physics, Mathematics