Skip to content
/ FDTD3D Public

A parallel implementation of 3D FDTD simulation using MPI.

License

Notifications You must be signed in to change notification settings

soloice/FDTD3D

Folders and files

NameName
Last commit message
Last commit date

Latest commit

6e004d9 · Sep 11, 2014

History

11 Commits
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 24, 2014
Aug 24, 2014
Sep 11, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014
Aug 24, 2014
Aug 26, 2014
Aug 24, 2014
Aug 24, 2014
Aug 24, 2014
Aug 26, 2014
Aug 24, 2014
Aug 26, 2014
Aug 26, 2014
Aug 26, 2014

Repository files navigation

What is FDTD?

FDTD(Finite-difference time-domain) method is a numerical analysis technique used for modeling computational electrodynamics. The idea is rather simple, but this method involves a lot of computation, which makes it sometimes intolerably slow to run on a typical PC. As the development of parallel computation, we may run FDTD program on a distributed system. Though some business software has good implementation on this problem, it is really difficult to find an open-source implementation. What's more, the numerical simulation of the spread of electromagnetic field is a pretty basic problem, so that many researchers may want to use it. Therefore, we decided to work on this problem and publish our code.

Our code turns to have a good scalability. Experiments show that we achieves an 6X speedup on a cluster consisting of 8 computing nodes.

We recommand you to run this code on a Linux cluster(or a single PC) with MPI environment.

Author:

  • Chong Ruan: OpenMP & MPI parallelization, inter-process communication.
  • Weiqiang Zhu: serial program implementaion, generating test data and result visualization.

Hint:

  • All .cpp files are main routines
  • All .in files are input data
  • All .dat files are output data
  • All .m or .py files are used for debug and data visualization

Feel free to report any bugs to ruanchong_ruby@163.com. We wish you a good luck!

About

A parallel implementation of 3D FDTD simulation using MPI.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published