View Stash data using the web
Stash allows you to access files using your web browser. In order to do
this, you'll need to put your file in
~/data/public (the two locations
point to the same directory). Any file or directory that is placed
here will be made available in the Stash webserver. Let's make a file
available using the Stash webserver:
$ cd ~/public $ echo "This is served over the web" > web-file
Now go to
http://stash.osgconnect.net/+username/ in your browser. You should
see the file in the listing. Clicking on the file should give you the contents.
Using data on Stash in compute jobs
Let us do an example calculation to understand the use of Stash and how we download
the data from the web. We will peform a molecular dynamics simulation of a small
protein in implicit water. To get the necessary files, we use the
tutorial command on
Log in to OSG Connect:
$ ssh firstname.lastname@example.org
$ tutorial stash-namd $ cd ~/tutorial-stash-namd
Aside: NAMD is a widely used molecular dynamics simulation program. It lets users specify a molecule in some initial state and then observe its time evolution subject to forces. Essentially, it lets you go from a specifed molecular structure to a simulation of its behavior in a particular environment. It has been used to study polio eradication, similations of graphene, and studies of biofuels.
You should see the following files in the directory:
$ ls namd_stash_run.submit #HTCondor job submission script file. namd_stash_run.sh #Job execution script file. ubq_gbis_eq.conf #Input configuration for NAMD. ubq.pdb #Input pdb file for NAMD. ubq.psf #Input file for NAMD. par_all27_prot_lipid.inp #Parameter file for NAMD.
par_all27_prot_lipid.inp is the parameter file and is required for
the NAMD simulations. The parameter file is common data file for the NAMD
mv par_all27_prot_lipid.inp ~/public/.
You can view the parameter file using your web browser by going to
Now we want the parameter file available on the execution (worker) machine when the simulation starts to run. As mentioned earlier, the data on the Stash is available to the execution machines. This means the execution machine can transfer the data from Stash as a part of the job execution. So we have to script this in the job execution script.
You can see that the job script
namd_stash_run.sh has the following lines:
$ cat namd_stash_run.sh #!/bin/bash source /cvmfs/oasis.opensciencegrid.org/osg/modules/lmod/5.6.2/init/bash module load namd/2.9 wget http://stash.osgconnect.net/+username/par_all27_prot_lipid.inp namd2 ubq_gbis_eq.conf
In the above script, you will have to insert your
username in URL address. The
parameter file located on Stash is downloaded using the #wget# utility.
Now we submit the NAMD job:
$ condor_submit namd_stash_run.submit
Once the job completes, you will see a non-empty
job.out.0 file, the standard output (
stdout) from the job.
$ tail job.out.0 WallClock: 6.084453 CPUTime: 6.084453 Memory: 53.500000 MB Program finished.
The above lines indicate the NAMD simulation was successful.
- [x] Your Stash data is located at
- [x] Data on Stash can be accessed by jobs running on compute nodes in the OSG.
This page was updated on Oct 23, 2019 at 11:00 from tutorials/tutorial-stash-namd/README.md.