Masonry example without fully postproces
Re: Masonry example without fully postproces
Thanks!
But why the data in the second file didn't contain the full-time history? It should have all the data during the 13400s instead of only the 450s.
Looking forward to your reply!
But why the data in the second file didn't contain the full-time history? It should have all the data during the 13400s instead of only the 450s.
Looking forward to your reply!
Re: Masonry example without fully postproces
Sorry but I cannot test it, there is a tcl file for custom recorders that is not in the attachments
Re: Masonry example without fully postproces
Hello, you can just remove it.
It's just a 'recorder' file with code below:
----------------------------------------
set filename Results
file mkdir $filename
recorder Element -file $filename/ele-20-localforce.txt -time -ele 20 localForce;
recorder Node -file $filename/nodes-4-disp.txt -time -node 4 -dof 2 disp;
recorder Node -file $filename/nodes-5-force.txt -time -node 5 -dof 2 reaction;
------------------------------------------
It's just a 'recorder' file with code below:
----------------------------------------
set filename Results
file mkdir $filename
recorder Element -file $filename/ele-20-localforce.txt -time -ele 20 localForce;
recorder Node -file $filename/nodes-4-disp.txt -time -node 4 -dof 2 disp;
recorder Node -file $filename/nodes-5-force.txt -time -node 5 -dof 2 reaction;
------------------------------------------
Re: Masonry example without fully postproces
In my case it works fine.
However, since it's a parallel analysis with opensees mp, you need a parallel numberer and the mumps solver (see the attached file)
However, since it's a parallel analysis with opensees mp, you need a parallel numberer and the mumps solver (see the attached file)
Re: Masonry example without fully postproces
I get it!
Thanks for your reply!
Thanks for your reply!
Re: Masonry example without fully postproces
You're welcome
Re: Masonry example without fully postproces
Dear STKOteam,
After using your recommended method for some time, I hope you can come up with a more convenient way to record node data.
Your proposed method creates only N files for N processes, and each file contains data for many nodes, making it difficult to find out which node the data belongs to.
Is there a way to output all of a node's data into a separate file under parallel computing?
After using your recommended method for some time, I hope you can come up with a more convenient way to record node data.
Your proposed method creates only N files for N processes, and each file contains data for many nodes, making it difficult to find out which node the data belongs to.
Is there a way to output all of a node's data into a separate file under parallel computing?
STKO Team wrote: ↑Mon Nov 20, 2023 9:43 amOtherwise, if you want to use plain recorders, you must create them one per process. Something like this:
recorder Node -file "ANodeRecorder_[getPID].txt" ....
where the [getPID] function will return the ID of the current processor.
So, in this way you will have N files, one for each process.
Then you can do some scripting to merge them
Re: Masonry example without fully postproces
Simple text-based node recorders are in OpenSees, STKO has no control over them.
Why don't you use the Monitor function in STKO? It's used to monitor a node response in real-time, but all it does is to write node data into a file, taking care of the processors.
Otherwise you can use the MPCORecorder and use the Python API to extract data from STKO post processor
Why don't you use the Monitor function in STKO? It's used to monitor a node response in real-time, but all it does is to write node data into a file, taking care of the processors.
Otherwise you can use the MPCORecorder and use the Python API to extract data from STKO post processor