Difference between revisions of "MEEP"

From EM Drive
Jump to: navigation, search
m (Setup)
(Added instructions from @leomillert)
Line 4: Line 4:
 
Details regarding setting up Meep for EM Drive simulations are below:
 
Details regarding setting up Meep for EM Drive simulations are below:
  
= Setup =
+
= Linux Setup =
 
The following was recommended by @leomillert:<ref>[http://forum.nasaspaceflight.com/index.php?topic=37642.msg1404656#msg1404656 Forum post by @leomillert]</ref>
 
The following was recommended by @leomillert:<ref>[http://forum.nasaspaceflight.com/index.php?topic=37642.msg1404656#msg1404656 Forum post by @leomillert]</ref>
* Linux
 
 
* [https://github.com/stevengj/meep MEEP 1.3] with libctl 3.2.2, Guile 2.0.11, Harminv 1.4 and OpenBLAS 0.2.12.  HDF5 is version 1.9.224 and h5utils is 1.12.1.  I installed everything except OpenBLAS from source, compiled myself. It was really tricky, so I strongly recommend people to see first if their distribution has MEEP pre-packaged.
 
* [https://github.com/stevengj/meep MEEP 1.3] with libctl 3.2.2, Guile 2.0.11, Harminv 1.4 and OpenBLAS 0.2.12.  HDF5 is version 1.9.224 and h5utils is 1.12.1.  I installed everything except OpenBLAS from source, compiled myself. It was really tricky, so I strongly recommend people to see first if their distribution has MEEP pre-packaged.
* [http://ab-initio.mit.edu/wiki/index.php/Meep_Installation Installation sequence]  
+
* The basic installation sequence for a single machine is documented [http://ab-initio.mit.edu/wiki/index.php/Meep_Installation here]Parallel / cluster installs are documented [http://ab-initio.mit.edu/wiki/index.php/Meep_Installation#MPI_.28parallel_machines.29 here]
* Parallel / cluster installs are documented [http://ab-initio.mit.edu/wiki/index.php/Meep_Installation#MPI_.28parallel_machines.29 here]
 
  
== Amazon EC2 AMI ==
+
= Usage =
 +
These instructions were compiled by @leomillert:<ref>[https://forum.nasaspaceflight.com/index.php?topic=37642.msg1404977#msg1404977 Forum post by @leomillert]</ref>
 +
 
 +
1. Execute
 +
<pre>
 +
wget https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=37642.0;attach=1042821 -O NSF-1701.ctl
 +
meep NSF-1701.ctl
 +
</pre>
 +
2. Eventually, MEEP will output nine .h5 files. It may take a long time depending on your computer. Patience is a virtue.
 +
 
 +
3. Create the CSV files:
 +
<pre>
 +
h5totxt -t 13 -0 -y -0  ex.h5 > zCopper-exy.csv
 +
</pre>
 +
4.  Open your zCopper-exy.csv on a spread sheet and aero's zCopper-exy.csv on another. Open a third spread sheet that is one spread sheet minus the other, entry by entry. Check that highest entry (in absolute value). If it's negligible you are good to go. If it's a value too big, greater than 10^-6, your MEEP installation isn't in sync with ours, so it's no use.
 +
 
 +
5. Now you are good to go. Make a new directory to start the tests. Copy NSF-1701.ctl there.
 +
 
 +
6. Open NSF-1701.ctl in a text editor and change a single value. For example, (set! high 10.2) means the model is 10.2 inches high. Change the 10.2 to another value and save NSF-1701.ctl with this single change. This is called sensitivity analysis. One value at a time. (set! high 10.2) was just an example, change any value of interest
 +
<pre>
 +
meep NSF-1701.ctl
 +
h5totxt -t 13 -0 -y -0  ex.h5 > zCopper-exy.csv
 +
</pre>
 +
7. Compare your new zCopper-exy.csv with your old one. See if there was any relevant change (do the spreadsheet comparison again). If there was no considerable change in values, it means the modification made doesn't impact the behavior of the EMDrive. This is an important information for scientists, so let us know. Otherwise, if there was a significant change, let us know if it was positive or negative and its intensity. If you don't know how, just upload the .h5 files somewhere and we will analyze it.
 +
 
 +
= Amazon EC2 AMI =
 
A public AMI with meep installed has been made available (by @dumbo). The AMI name is ubuntu-trusty-meep-emdrive and the AMI ID is ami-3f54560f. Due to the long computing times required to simulate the frustum until "saturation" (for lack of a better term), some people may find it more convenient to run the simulation on EC2 than on their local workstation.
 
A public AMI with meep installed has been made available (by @dumbo). The AMI name is ubuntu-trusty-meep-emdrive and the AMI ID is ami-3f54560f. Due to the long computing times required to simulate the frustum until "saturation" (for lack of a better term), some people may find it more convenient to run the simulation on EC2 than on their local workstation.
  

Revision as of 07:44, 15 July 2015

Meep (or MEEP) is a free finite-difference time-domain [1] simulation software package developed at MIT to model electromagnetic systems. A number of people, including @aero, have been using it to simulate the evolution of the fields within the EM Drive.

Details regarding setting up Meep for EM Drive simulations are below:

Linux Setup

The following was recommended by @leomillert:[2]

  • MEEP 1.3 with libctl 3.2.2, Guile 2.0.11, Harminv 1.4 and OpenBLAS 0.2.12. HDF5 is version 1.9.224 and h5utils is 1.12.1. I installed everything except OpenBLAS from source, compiled myself. It was really tricky, so I strongly recommend people to see first if their distribution has MEEP pre-packaged.
  • The basic installation sequence for a single machine is documented here. Parallel / cluster installs are documented here

Usage

These instructions were compiled by @leomillert:[3]

1. Execute

wget https://forum.nasaspaceflight.com/index.php?action=dlattach;topic=37642.0;attach=1042821 -O NSF-1701.ctl
meep NSF-1701.ctl

2. Eventually, MEEP will output nine .h5 files. It may take a long time depending on your computer. Patience is a virtue.

3. Create the CSV files:

h5totxt -t 13 -0 -y -0  ex.h5 > zCopper-exy.csv

4. Open your zCopper-exy.csv on a spread sheet and aero's zCopper-exy.csv on another. Open a third spread sheet that is one spread sheet minus the other, entry by entry. Check that highest entry (in absolute value). If it's negligible you are good to go. If it's a value too big, greater than 10^-6, your MEEP installation isn't in sync with ours, so it's no use.

5. Now you are good to go. Make a new directory to start the tests. Copy NSF-1701.ctl there.

6. Open NSF-1701.ctl in a text editor and change a single value. For example, (set! high 10.2) means the model is 10.2 inches high. Change the 10.2 to another value and save NSF-1701.ctl with this single change. This is called sensitivity analysis. One value at a time. (set! high 10.2) was just an example, change any value of interest

meep NSF-1701.ctl
h5totxt -t 13 -0 -y -0  ex.h5 > zCopper-exy.csv

7. Compare your new zCopper-exy.csv with your old one. See if there was any relevant change (do the spreadsheet comparison again). If there was no considerable change in values, it means the modification made doesn't impact the behavior of the EMDrive. This is an important information for scientists, so let us know. Otherwise, if there was a significant change, let us know if it was positive or negative and its intensity. If you don't know how, just upload the .h5 files somewhere and we will analyze it.

Amazon EC2 AMI

A public AMI with meep installed has been made available (by @dumbo). The AMI name is ubuntu-trusty-meep-emdrive and the AMI ID is ami-3f54560f. Due to the long computing times required to simulate the frustum until "saturation" (for lack of a better term), some people may find it more convenient to run the simulation on EC2 than on their local workstation.

The AMI is based on Ubuntu (Trusty) with packages meep and meep-mpi-default installed.

Models & results

TBD: Add reference to all existing models and the results here, with a succinct description.