Introduction

Running FORTRAN 77 programs on the cluster is similar to running any other serial job. Make sure you've read the tutorial for C programs first, to understand the basics.

Serial Example

Let's try to compile this simple Fortran 90 program

      write(*,*) "Salutations, Earth."
      end

Download: ../code/hello_serial-f77/hello_serial.f

To compile this using GCC and create the executable helloworld-f77-gcc, type

[araim1@tara-fe1 hello_serial-f77]$ gfortran -ffixed-form hello_serial.f -o hello_serial-f77-gcc
[araim1@tara-fe1 hello_serial-f77]$

Alternatively, to compile using PGI Fortran and create the executable helloworld-f77-pgi, type

[araim1@tara-fe1 hello_serial-f77]$ pgf77 hello_serial.f -o hello_serial-f77-pgi
[araim1@tara-fe1 hello_serial-f77]$

Running this program should produce this message:

[araim1@tara-fe1 hello_serial-f77]$ ./hello_serial-f77-gcc
Salutations, Earth.
[araim1@tara-fe1 hello_serial-f77]$ ./hello_serial-f77-pgi
Salutations, Earth.

Parallel Example

We'll assume that you already know the fundamentals of MPI execution from the C tutorial. Now let's create the following program:

      program hello_parallel

C     Include the MPI library definitons:                                                                                
      include 'mpif.h'

      integer numtasks, rank, ierr, rc, len, i
      character*(MPI_MAX_PROCESSOR_NAME) name

C     Initialize the MPI library:                                                                                        
      call MPI_INIT(ierr)
      if (ierr .ne. MPI_SUCCESS) then
         print *,'Error starting MPI program. Terminating.'
         call MPI_ABORT(MPI_COMM_WORLD, rc, ierr)
      end if

C     Get the number of processors this job is using:                                                                    
      call MPI_COMM_SIZE(MPI_COMM_WORLD, numtasks, ierr)

C     Get the rank of the processor this thread is running on.  (Each                                                    
C     processor has a unique rank.)                                                                                      
      call MPI_COMM_RANK(MPI_COMM_WORLD, rank, ierr)

C     Get the name of this processor (usually the hostname)                                                              
      call MPI_GET_PROCESSOR_NAME(name, len, ierr)
      if (ierr .ne. MPI_SUCCESS) then
         print *,'Error getting processor name. Terminating.'
         call MPI_ABORT(MPI_COMM_WORLD, rc, ierr)
      end if

      print 30, "hello_parallel.f: Number of",numtasks,rank,name
 30   format (A,' tasks=',I3,' My rank=',I3,' My name=',A80,'')
C     Tell the MPI library to release all resources it is using:                                                         
      call MPI_FINALIZE(ierr)

      end

Download: ../code/hello_parallel-f77/hello_parallel.f

Compile your program with

mpif77 hello_parallel.f -o hello_parallel

The same compilation command should work even if you've changed the switcher to use a different MPI implementation and compiler. Running the compiled program is now exactly the same as running a C MPI program.