CARC web archive · Archived copy of hpc.unm.edu/computing/roadrunner/ug/index.html as captured by the Internet Archive on 9 September 2003. Links to pages that were not saved here go to the Wayback Machine. Wayback copy · All pages in this snapshot · Web archive index
UNM-HPCERC: Linux Supercluster, Contents  

Alliance Linux Cluster User Modules

The Roadrunner Linux Cluster is a resource provided by funding to the AHPCC from the National Computational Science Alliance (Alliance). Approximately 85% of the available time on Roadrunner is reserved for Alliance-allocated users.

Table of Contents

  1. Architecture and System Configurations
  2. Allocations and Gaining Access
    1. Who may access Roadrunner
    2. Applying for an allocation
      1. Small or startup allocations
      2. Peer-reviewed allocations
      3. Award notification and default passwords
      4. AHPCC guest accounts
    3. Security and connecting to the cluster
      1. Eliminating clear text passwords
      2. Software for connecting securely
      3. Secure file transfer
      4. Reporting security incidents
      5. For more information
  3. Account Management
    1. Managing your account
      1. Changing your password
      2. What if I have forgotten my password?
      3. Changing your default shell
      4. Managing your "dot" (.cshrc or .bash_profile) file
      5. Forwarding your mail
    2. Managing your allocation
      1. Adding users to your allocation
      2. Removing users from your allocation
      3. Monitoring usage of your allocation
      4. Renewing your allocation
      5. Charging algorithm
  4. File Systems and Storage
    1. Directories
      1. Home directories
      2. Global scratch directories
      3. Local scratch directories
      4. /tmp and /usr/tmp directories
    2. Permanent file storage
    3. Backups and file restoration
  5. Compiling and Linking Codes
    1. Overview
      1. Available programming languages
      2. Myrinet versus Ethernet
      3. Available compiler suites
    2. Available compilers
      1. Fortran77
      2. Fortran90/95
      3. High Performance Fortran (HPF)
      4. C
      5. C++
    3. MPI codes
      1. Selecting an MPI library
        1. Settings for MPIHOME
        2. Setting MPIHOME in your dot files
        3. Setting MPIHOME interactively
      2. Compiling and linking with MPI
    4. Standard libraries
  6. Jobs
    1. Overview
    2. Interactive jobs
      1. Overview
      2. Requesting nodes for interactive jobs
      3. Running Myrinet interactive jobs
      4. Running Ethernet interactive jobs
    3. Batch jobs
      1. Basic LRM batch script
        1. Sample LRM batch script
      2. Submitting a LRM batch job
      3. Monitoring your job
      4. Monitoring the queue
      5. Monitoring the nodes
      6. Removing your job from the queue
      7. Redirecting from stdin
      8. More advanced LRM scripts
      9. LRM commands
      10. Using PBS scripts to submit your job
  7. Debugging and Optimization
    1. Available debuggers
      1. GNU gdb
        1. X Windows interface
      2. PGI pgdbg
        1. X Windows interface
      3. Etnus TotalView for MPI
    2. Available profilers
      1. PGI pgprof
  8. Resources and getting help
    1. Getting help
    2. Training
  9. Index