F90_CONDOR is a directory of examples which demonstrate how a FORTRAN90 program can be executed on a remote machine using the CONDOR batch queueing system.
CONDOR allows a user to submit jobs for batch execution on an informal cluster composed of various computers that often have idle time. Based on information from the user's submission file, CONDOR chooses one or more appropriate and available computers, transfers files to the target systems, executes the program, and returns data to the user.
CONDOR has many features, and its proper use varies from site to site. The information in this document was inspired by the CONDOR system supported by the FSU Research Computing Center (RCC). Some of the information therefore is peculiar to this local installation.
The first thing to note is that the FSU CONDOR cluster does not have a shared file system. Thus, you should probably imagine that you are trying to run your program on a remote machine that might be someone's iPhone. None of your files are there, and if you don't say so, you don't even know what kind of processor or operating system you are dealing with!
So you are responsible for explaining to CONDOR what files you want to transfer to this remote machine, and you must specify whether you have any requirements, such as a specific operating system or processor type. In particular, when working with a FORTRAN90 program, you must precompile the program on the CONDOR login node, and one of your requirements will then be that the remote computer has the same processor and operating system as the login node.
The information about file transfer and system requirements is all part of what is called a CONDOR submission script. The script is a combination of comment lines, which begin with the "#" sign, commands of the form "A = B" which set certain options, and a few commands that make things happen. A typical script might be called "program.condor".
The user and CONDOR interact on a special computer called the CONDOR login node or submit node. The user's files must be transferred there, the user's CONDOR submission script must be submitted there, and the results from the remote machine will be copied back to this place.
The user logs into the CONDOR submit node interactively:
ssh condor-login.rcc.fsu.edu
There are TWO reasons why this might not work for you, however.
File transfer is done with the SFTP command.
sftp condor-login.rcc.fsu.edu
put program.condor
put mandelbrot.f90
quit
When ready, the CONDOR submission script is sent by the command
condor_submit program.condor
The user can check on the status of the job with the command
condor_q
If all goes well, the job output will be returned to the CONDOR
submit node.
However, if things do not go well, or the job is taking too much
time, user "username" can delete all jobs in the condor queue with
the command
condor_rm username
On the FSU RCC Condor cluster, you must first copy your files to the CONDOR login machine. When you submit your job to the CONDOR queue, however, the program execution will take place on some unknown machine, which initially does not have any of your files. Therefore, an important part of using CONDOR is making sure that you copy to the remote machine all the files needed for input, including a compiled copy of your FORTRAN90 program. Luckily, CONDOR will automatically copy back to the login node every file that is created by the program execution.
Because the file system is not shared, the following commands should appear in your CONDOR script:
should_transfer_files = yes
when_to_transfer = on_exit
that allows you to specify the name of this file.
If your executable reads from "standard input", (such as the FORTRAN statement
READ ( *, * ) N
then your CONDOR job will need a file containing that information.
CONDOR includes a command of the form
input = filename
that allows you to specify the name of this file. Similarly, if
your program writes to "standard output", such as the FORTRAN90 statement
WRITE ( *, * ) 'The answer is ', 42
then CONDOR allows you to
specify the name of a file where this information will go:
output = filename
The input file must exist on your CONDOR login node before you submit
the job. The output file is created during the run, and
will automatically be copied back to your CONDOR login node when the
job is completed.
Your job may require other data files to run than simply the standard input file. If so, you need to tell CONDOR the names of these files, in a comma-separated list:
transfer_input_files = file1, file2, ..., file99
Your job may create many files aside from simply standard output. Luckily, all files created by the run will be automatically copied back.
We are going to compile the FORTRAN90 executable on the CONDOR login node, but run it on some other remote machine. Therefore, we need the following command to guarantee that the remote machine can understand the executable:
requirements = ( OpSYS="LINUX" && Arch=="X86_64 )
Suppose that the FORTRAN90 program we want to run is named "mandelbrot.f90". We compile this program on the CONDOR login node with commands like
gfortran mandelbrot.f90
mv a.out mandelbrot
which creates an executable program called "mandelbrot".
This program must be copied to the remote machine,
so your CONDOR script must include the statement:
executable = mandelbrot
Here is a file called "mandelbrot.condor" which ought to be able to run our compiled program somewhere, and return the results to us:
universe = vanilla
executable = mandelbrot
arguments =
input =
requirements = ( OpSYS="LINUX" && Arch=="X86_64 )
should_transfer_files = yes
when_to_transfer_files = on_exit
notification = never
output = mandelbrot.txt
log = mandelbrot_log.txt
queue
A few comments are in order.
The computer code and data files made available on this web page are distributed under the GNU LGPL license.
F90_CONDOR is available in a C version and a C++ version and a FORTRAN90 version and a MATLAB version
C_CONDOR, C programs which illustrate how a C program can be run in batch mode using the condor queueing system.
C++_CONDOR, C++ programs which illustrate how a C++ program can be run in batch mode using the condor queueing system.
CONDOR, examples which demonstrates the use of the CONDOR queueing system to submit jobs that run on a one or more remote machines.
F77_CONDOR, FORTRAN77 programs which illustrate how a FORTRAN77 program can be run in batch mode using the condor queueing system.
MATLAB_CONDOR, programs which illustrate how a MATLAB program can be executed with the CONDOR batch queueing system.
MANDELBROT is an example which creates a PPM image file of the Mandelbrot set.
You can go up one level to the FORTRAN90 source codes.
