UCSD Logo UCSD Logo For Printing Skip navigation links

Navigation

UCSD Triton Resource @ SDSC

Quick Status

Triton Resource Node Status

Saturday, November 21st 2009 01:59:01 PM PST


TCC Rack 3 Nodes Down (1)

tcc-3-71.local

Total TCC Nodes Up: 247

Total 256GB (PDAF) Nodes Up: 20

Total 512GB (PDAFM) Nodes Up: 8

Rack 2 Up Count: 80

Rack 3 Up Count: 77

Rack 4 Up Count: 11

Rack 5 Up Count: 79

Compiling Parallel Codes on Triton Resource

This page shows simple examples of how to compile codes for the Tritan Resource. There is a C example and an equivalent Fortran example.

Compiling for Triton Resource


Note: The Triton Resource is now available to users in full production mode. Configuration and testing of Triton is complete. The Triton Compute Cluster (TCC) and Petascale Data Analysis Facility (PDAF) are using TAPP accounts to charge users for compute time as of Monday, October 5, 2009.

Early Adopter accounts have been converted to trial accounts and provisioned with 1000 complimentary SUs.

TAPP, the Triton Affiliates and Partners Program, is the prescribed way to manage your access.

Triton staff maintain a Discussion List to which all Triton users are encouraged to subscribe. Members can post questions and comments to Triton Discussion List (triton-discuss@sdsc.edu) to obtain help and support for issues and community feedback.

  1. Compiling
  2. The default compiler/mpi stack combination is the Portland compiler (pgf90, pgf77, pgcc, pgCC) and openmpi. Other compilers and mpi variants may be accessed by loading the appropriate modules. The commands mpicc, mpiCC, mpif77, mpif90 will access a particular compiler/mpi combination based on the module choices. Read about modules to learn how Triton Resource manages compiler configurations.

    A simple MPI C program is given below (mpi_c.c):

  3. C Example
  4. #include <stdio.h>
    #include <stdlib.h>
    #include <mpi.h>
    main(int argc,char *argv[])
    {
       int myproc,numproc;
       MPI_Status status;
    
       MPI_Init(&argc,&argv);
    
       MPI_Comm_rank(MPI_COMM_WORLD,&myproc);
       MPI_Comm_size(MPI_COMM_WORLD,&numproc);
    
       if(myproc == 0)
            printf("NUMPROC %d\n",numproc);
       printf("Process %d\n running",myproc);
       MPI_Finalize();
    }
    
    

    This program can be compiled with the following command

    • mpicc -o mpi_c mpic.c

    The program will print the total number of MPI processes initiated and a message printed by each process (see Submitting a job)

  5. Fortran Example
  6. Following is the Fortran equivalent of the above program (mpi_f.f):

    program mpi
    implicit double precision (a-h,o-z)
    include "mpif.h"
    
    call mpi_init(ierror)
    
    call mpi_comm_rank(MPI_COMM_WORLD,myproc,ierror)
    call mpi_comm_size(MPI_COMM_WORLD,numproc,ierror)
    if(myproc.eq.0) write(6,*) 'NUMPROC ',numproc
    write(6,*) 'Process ',myproc,' running'
    call mpi_finalize();
    end
    
    

    and can be compiled as follows :

    • mpif77 -o mpi_f mpi_f.f

    More information about compiling can be found on the Compiler Jobs page.

Contact Us

Open a Ticket with Triton Resource Support using the Support Ticket Form.

Join the Discussion Forum Sign up for our Email Discussion List.

Follow Triton on Twitter

FAQ Read the FAQ Page.

Terms of Use | Privacy