Send meep-discuss mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of meep-discuss digest..."
Today's Topics:
1. Slow execution issue (Andrea Armaroli)
2. Re: about magnetostatic waves (Matthew Miskiewicz)
3. Unit transformation of dispersion parameters (???)
4. Question about magnetostatic waves (Alexandr Sadovnikov)
----------------------------------------------------------------------
Message: 1
Date: Mon, 2 Aug 2010 18:38:57 +0200
From: Andrea Armaroli <[email protected]>
Subject: [Meep-discuss] Slow execution issue
To: [email protected]
Message-ID:
<[email protected]>
Content-Type: text/plain; charset="iso-8859-1"
Dear MEEP users, prof. Johnson, and other developers,
I'm facing an execution speed issue, that is remarkably slowing down my
work.
I am trying to study the transmission/reflection response of a structure
made of semiconductor and metal films (5 poles for 2 media). Also the
absorption of complex epsilon media interests me.
My domain comprises
(451, 376, 120)
points, and I set 20 as resolution, 0.3 as Courant constant.
Lauched with run-sources+ 5000 (it should be about 350000 steps.)
My choice of using a smaller Courant number than usual is due to divergence
of calculation in a previous attempt.
Well, the structure without scatterers (only dielectric media) keeps running
for 3 days on 6 nodes (1 processor per node) of our (brand new) cluster.
The structure with scatterers has run for 4 days and has completed only half
of the task.
The first run wrote epsilon, a single point probe, and the flux to subtract
in the second run. This latter is slow even without writing the probe file.
We compiled the code with openMPI on Scientific Linux.
I though serial HDF was the problem, but a previous discussion claims
parallel HDF is worse unless a huge number of CPUs is used.
My boss is an FDTD expert and we're comparing MEEP with his own
implementation. He is a bit surprised by such a long execution and we would
like to understand why. Could the double precision account for that? Or what
else.
Thank you very much for your help and suggestions.
Andrea.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://ab-initio.mit.edu/pipermail/meep-discuss/attachments/20100802/0cbe2cca/attachment.html>
------------------------------
Message: 2
Date: Mon, 2 Aug 2010 15:01:10 -0400
From: Matthew Miskiewicz <[email protected]>
Subject: Re: [Meep-discuss] about magnetostatic waves
To: [email protected]
Message-ID:
<[email protected]>
Content-Type: text/plain; charset="iso-8859-1"
Hi Jian,
The problem is this line:
(set! default-material (make material-function (epsilon-func feps)))
Since you are using epsilon-func and not material-func, you must specify an
actual material. Changing the line to this makes it work:
(set! default-material (make dielectric (epsilon-func feps)))
Cheers,
Matt Miskieiwcz
On Mon, Aug 2, 2010 at 11:21 AM, zhang jian <[email protected]> wrote:
Hi, dear Meep users, I am a beginner of Meep. I write a simple meep
code, and run it at the server of my department. However it can not
calculate a final result, though it is just a simple code. Could you
please tell me how to simply this code? Thank you very much.~
Initializing structure...
Working in 2D dimensions.
block, center = (-5,5,0)
size (10,1,1e+20)
axes (1,0,0), (0,1,0), (0,0,1)
dielectric constant epsilon = 11.9
block, center = (2,6.5,0)
size (10,1,1e+20)
axes (1,0,0), (0,1,0), (0,0,1)
dielectric constant epsilon = 11.9
subpixel-averaging is 19.964% done, 18.708 s remaining
subpixel-averaging is 39.948% done, 7.03004 s remaining
subpixel-averaging is 59.9321% done, 3.10042 s remaining
subpixel-averaging is 79.9161% done, 1.16184 s remaining
subpixel-averaging is 99.9001% done, 0.0046283 s remaining
subpixel-averaging is 19.964% done, 18.5258 s remaining
subpixel-averaging is 39.948% done, 6.83587 s remaining
subpixel-averaging is 59.9321% done, 3.094 s remaining
subpixel-averaging is 79.9161% done, 1.16121 s remaining
subpixel-averaging is 99.9001% done, 0.00463442 s remaining
subpixel-averaging is 13.9447% done, 28.105 s remaining
subpixel-averaging is 27.9034% done, 10.931 s remaining
subpixel-averaging is 41.8621% done, 6.41271 s remaining
subpixel-averaging is 55.8208% done, 3.39532 s remaining
subpixel-averaging is 69.7795% done, 1.92881 s remaining
subpixel-averaging is 83.7381% done, 0.868258 s remaining
subpixel-averaging is 97.6968% done, 0.102055 s remaining
subpixel-averaging is 13.9447% done, 28.8167 s remaining
subpixel-averaging is 27.9034% done, 12.0302 s remaining
subpixel-averaging is 41.8621% done, 6.41883 s remaining
subpixel-averaging is 55.8208% done, 3.66837 s remaining
subpixel-averaging is 69.7795% done, 2.00258 s remaining
subpixel-averaging is 83.7381% done, 0.89728 s remaining
subpixel-averaging is 97.6968% done, 0.109034 s remaining
subpixel-averaging is 1.20386% done, 359.114 s remaining
subpixel-averaging is 2.40893% done, 177.411 s remaining
subpixel-averaging is 3.61399% done, 115.726 s remaining
subpixel-averaging is 4.81906% done, 85.4739 s remaining
subpixel-averaging is 6.02413% done, 66.8641 s remaining
subpixel-averaging is 7.22919% done, 55.6286 s remaining
..........................................
My code:
(reset-meep)
(define eps-si 11.9)
(define eps-sio2 (expt 1.46 2))
(define wr 1)
(define bendr 5)
(define bend-xcen 0)
(define bend-ycen 0)
(define gap 0.5)
(define (circ p) (
+ (expt (- (vector3-x p) bend-xcen) 2) (expt (-
(vector3-y p) bend-ycen) 2)
)
)
(define (fline1 p) (- (vector3-x p) bend-xcen))
(define (fline2 p) (- (vector3-y p) bend-ycen))
(define r1 (+ bendr (/ wr 2)))
(define r2 (- bendr (/ wr 2)))
(define (feps p)
(if
(and (<= (circ p) (expt r1 2))
(>= (circ p) (expt r2 2))
(>= (fline1 p) 0)
(>= (fline2 p) 0)
)
eps-si eps-sio2
)
)
(set! geometry-lattice (make lattice (size 22 16 no-size)))
(set! default-material (make material-function (epsilon-func feps)))
(set! geometry
(append
(list
(make block (center -5 bendr) (size 10 wr infinity)
(material (make dielectric (epsilon eps-si))))
(make block (center 2 (+ bendr (+ wr gap))) (size 10 wr
infinity)
(material (make dielectric (epsilon eps-si))))
))
)
(set! pml-layers (list (make pml (thickness 1.0))))
(set! resolution 10)
(set! sources (list
(make source
(src (make continuous-src
(wavelength (* 2 (sqrt 12))) (width 20)))
(component Ez)
(center -10 bendr) (size 0 1))))
(run-until 200
(at-beginning output-epsilon)
(to-appended "pbs" (at-every 0.6 output-efield-z)))
Jian
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://ab-initio.mit.edu/pipermail/meep-discuss/attachments/20100802/17602567/attachment-0001.htm>
------------------------------
Message: 3
Date: Tue, 3 Aug 2010 10:21:50 +0800
From: ??? <[email protected]>
Subject: [Meep-discuss] Unit transformation of dispersion parameters
To: [email protected]
Message-ID:
<[email protected]>
Content-Type: text/plain; charset="windows-1252"
Dear all
I have a question about the content of ?Meep Tutorial/Material dispersion?.
http://ab-initio.mit.edu/wiki/index.php?title=Meep_Tutorial/Material_dispersion&oldid=3762
There are two dielectric functions listed there, one is a function of w,
another is a function of f. As my realization, w=f in Meep, ex: f=1 then
w=1. The two functions should be identical. Why is the gamma divided by 2*pi
in function of f? If gamma needs this adjustment, omega has the same unit
with gamma, why is it not divided by 2*pi?
I am confused about unit transformation. If I find a dispersion relation in
SI unit as:
epsilon(f)= W + (X*Y^2) / (Y^2 - f^2 - i*f*Z) (SI unit)
f is the frequency (Hertz)
W, X (No unit)
Y, Z, (Hertz)
I define Meep unit length = a (meter), speed of light c = 3e8 (m/s)
The frequency unit is c/a (Hertz). So I define the parameters in Meep code
as
epsilon W
omega (/ Y (/ c a))
gamma (/ Z (/ c a))
sigma X
Is the transformation correct?
Any help will be appreciated.
Thanks a lot!
Wen-Yu Chen
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://ab-initio.mit.edu/pipermail/meep-discuss/attachments/20100803/048c80e7/attachment-0001.htm>
------------------------------
Message: 4
Date: Tue, 03 Aug 2010 16:01:48 +0400
From: Alexandr Sadovnikov <[email protected]>
Subject: [Meep-discuss] Question about magnetostatic waves
To: [email protected]
Message-ID: <[email protected]>
Content-Type: text/plain; charset=koi8-r
Hello, dear Prof. Steven and Meep Users.
I would like to ask you about application and use of
meep to the computation of magnet systems such as
ferromagnetic films in which the magnetostatic waves can
propagate. So I'm going to do this computation (of magnetostatic
wave propagation) using meep and ctl(scheem, aka libctl) interface.
The problem is that in this computation one have to use
the relative permeability tensor (aka mu) and
the components of this tensor depend on the frequency.
I suppose, that one have to rewrite some part of meep code
and recompile to do this. Maybe this problem can be solved
if we only change (in our imagination) that "eps - mu"
and will compute in this way. But I don't know unfortunately
how can one use in computation the tensor of relative permittivity
or permeability which components depend on frequency. That is the
trouble for me.
Thank you in advance for any possibly help.
Alexandr.
Postgraduated student. Saratov State University. Russia.
[email protected]
------------------------------
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
End of meep-discuss Digest, Vol 54, Issue 2
*******************************************
