Difference between revisions of "Running UCVM on Discovery"
Line 88: | Line 88: | ||
On discovery, setup your plotting environment. This is only needed the first time | On discovery, setup your plotting environment. This is only needed the first time | ||
− | %mkdir / | + | Confirm your test directory exists, if not create it with mkdir command |
+ | |||
+ | %mkdir /home1/maechlin/ucvm_test | ||
Then add these aliases into .bashrc file | Then add these aliases into .bashrc file | ||
− | "alias 2test='cd / | + | "alias 2test='cd /home1/maechlin/ucvm_test'" |
− | "alias 2project='cd /project/ | + | "alias 2project='cd /project/maechlin_162/ucvm/ucvm_bin'" |
Save the updated .bashrc file | Save the updated .bashrc file |
Revision as of 17:21, 31 March 2021
Contents
CARC Getting Started Documentation
Running ucvm_query on Discovery
This is the contents of the README.md on discovery
(base) [maechlin@discovery1 test_ucvm]$ cat README.md == Linux Command Syntax == Example commands on Linux system. Don't type the % sign %ls %pwd == ucvm query Input format == lon, lat, depth -118.0, 34.0 0.0 == Example query format == %ucvm_query -f /project/maechlin_162/ucvm_bin/conf/ucvm.conf -m cvmh < test_pts.in == allocation interactive nodes == %salloc --ntasks=1 --time=1:00:00 --account=scec_608 == Example profile query == %ucvm_query -f /project/maechlin_162/ucvm_bin/conf/ucvm.conf -m cvmsi < rpv.in > rpv_cvmsi.out == slurm batch script == %cat ucvm_query.job then.. %sbatch ucvm_query.job then .. %squeue -u maechlin then .. %cat rpv_cvmsi.out == UCVM Basin Query == *[https://github.com/SCECcode/UCVMC/wiki/Test-Cases-Basin-Query Basin Query Examples] == Example basin_query Command on Discovery == %basin_query -f /project/maechlin_162/ucvm_bin/conf/ucvm.conf -m cvmh -v 2500 < basin_pts.in > basin_pts_cvmh.out then.. %cat basin_pts_cvmh.out which shows returns 3 depth values tuple that is, first-crossing, second crossing and the last crossing with the threshold. == Transfer files between Discovery and laptop == Assume file on laptop, copy it to directory on discovery Issue this command from directory on laptop that contains the file ucvmPlotting.yml that you want to copy %scp ucvmPlotting.yml maechlin@discovery.usc.edu:/home1/maechlin/test_ucvm/. or, copy files from directory on discovery.usc.edu to laptop in a directory on laptop, run this command %scp maechlin@discovery.usc.edu:/home1/maechlin/test_ucvm/*
Create Plot using Website - 1D Vertical Profile
Create 1D vertical profile for location in LA Basin (CSU Dominquez Hills):
- 33.8628
- -118.2525
- 0-10k
- 100m
Create Plot using Website:
- point browser to http://moho.scec.org/UCVM_web/web/viewer.php
- select "1D Vertical Profile"
- enter 33.8628 for Latitude
- enter -118.2525 for Longitude
- enter 0 for Z start
- enter 10000 for Z ends
- enter 100 for Z step
Create Plot using Discovery
Assuiming plotting has been setup for your account, you can Setup the plotting environment,
%2test
then..
%plot_depth_profile.py -s 33.8628,-118.2525 -b 0 -e 10000 -d vs,vp,density -v 100 -c cvmh -o depth.png
Then, you can view the plots with
%view_png.py -f depth.png
Running UCVM Plotting on Discovery
On discovery, setup your plotting environment. This is only needed the first time
Confirm your test directory exists, if not create it with mkdir command
%mkdir /home1/maechlin/ucvm_test
Then add these aliases into .bashrc file
"alias 2test='cd /home1/maechlin/ucvm_test'"
"alias 2project='cd /project/maechlin_162/ucvm/ucvm_bin'"
Save the updated .bashrc file
then,..
%source ~/.bashrc
then,..
%env |grep UCVM
should see setting of UCVM related environment variable like these..
UCVM_SRC_PATH=...
UCVM_INSTALL_PATH=...
then, these command should return useful information..
%which ucvm_query
%ucvm_query -H
%conda env list
This alias should move you to your ucvm installation directory
%2project
Setting up Plotting Environment for the first time
if you see ucvmPlotting, do this, %conda activate ucvmPlotting skip to NEXT else, %conda env create -f ucvmPlotting.yml this will take a while.. %conda env list should see ucvmPlotting as one of the environment %conda activate ucvmPlotting %2project %git clone https://github.com/SCECcode/ucvm_plotting.git %cd ucvm_plotting %./unpack-dist NEXT: check if ucvm_plotting is there, %2project %cd ucvm_plotting/ucvm_plotting %./run_depth.sh to run manually, %plot_horizontal_slice.py -b 33.35,-118 -u 34.35,-117 -e 1000 -d vs -c cvmh -a s -s 0.01 -o horizontal.png %view_png.py -f horizontal.png