Page 1 of 1
OpenSees Lab on NEES hub
Posted: Wed Oct 05, 2011 10:04 am
by mkfotouhi
Frank,
1-I tried to run my model using OpenSees Hub and it took 15 min and 50 sec. When I use my own computer it takes 14 min 58 sec! Since it's supposed the NEES hub be very very very faster than normal PCs Do I need to add any command to my code or do I us the NEES hub tool wrongly?
2-I have still problem downloading OpenSees interpreter for parallel analyzing!
http://opensees.berkeley.edu/OpenSees/c ... .1.exe.zip
Thanks,
Re: OpenSees Lab on NEES hub
Posted: Thu Dec 15, 2011 12:44 pm
by fmk
sorry for the delay, i have been busy working on other things.
1) are you just running a regular sequential script?
2) working on the parallel version. not quite so easy this time as the machine on which i used to build it died and the parallel
version requires an awful lot of other libraries to be compiled and linked.
Re: OpenSees Lab on NEES hub
Posted: Thu Dec 15, 2011 1:45 pm
by mkfotouhi
Thanks Frank for your reply,
1) The only commands that I have in my scripts to account for the parallel analysis are as follows:
set parallel 1; # 0 -- sequential run (e.g., on PC)
# 1 -- parallel run
switch $parallel {
0 {
set solver "SparseSPD"
set outfmt "-file"
}
1 {
set solver "Mumps"
set outfmt "-xml"
}
}
2) thanks for your great job. It seems that this time it may take very long!
Re: OpenSees Lab on NEES hub
Posted: Thu Dec 15, 2011 1:47 pm
by fmk
are you outputting a lot of data?
Re: OpenSees Lab on NEES hub
Posted: Thu Dec 15, 2011 1:58 pm
by mkfotouhi
I don't know what's the scale but here are my recorder commands:
eval "recorder Node $outfmt NEESUndrainedTravelingTimeVerticalOnPile.dsp -time -node $SurfaceNodes $pileNodes -dof 1 2 3 disp"
eval "recorder Node $outfmt NEESUndrainedTravelingTimeVerticalOnPile.acc -time -node $SurfaceNodes $pileNodes -dof 1 2 3 accel"
The output is two files each around 3.5 MB
Re: OpenSees Lab on NEES hub
Posted: Mon Jan 23, 2012 10:16 am
by Dori
I seem to be having a similar issue, and suggestions? This is a bit frustrating. Thank you!