|
Computational
ElectroMagnetics
Theory-Algorithm-Code-Hardware
Our
goal is to develop systems simulations capabilities based on time-domain
computational electromagnetics methods. This ambitious goal gives
the design engineers the ability to simulate pieces of or entire systems
for electromagnetic interactions. In order to achieve the goal, several
fundamentals, which inherently provide or deny any capabilities the eventual
simulation tools will have, must be considered:
 |
The electromagnetics
theory (time-domain, frequency-domain, frequency-dependent, et
cetera) essentially dictates the possible capabilities of the electromagnetics
algorithm. |
 |
The programming
theory and algorithm (C, FORTRAN, C++, serial, parallel, MPI, threads,
hard-coded, static versus dynamic task assignment, et cetera) essentially
dictate the possible capabilities of the electromagnetics algorithm
and
the final simulations code. |
 |
The hardware
(disks, disk access methods in code, networking to disks, disk file systems,
CPU architecture, cache size/hierarchy, memory size/hierarchy, et cetera)
which must eventually perform the computations strongly influences each
of the aforementioned concerns and must be addressed during development
as an integral design component. |
CEM TACH emphasizes
the nature of our group which is grounded in Theory, Algorithm, Code and
Hardware development directed towards creating very capable, very general
and very fast simulation tools.
|