[an error occurred while processing this directive] [an error occurred while processing this directive]

Case and File Name Conventions


Case and file name conventions are used to help keep track of the numerous simulations and their output data. Three types of output are expected from each active component in a coupled run:
  • A series of large binary "history files" containing the gridpoint data for each component during the course of the run.
  • A series of smaller restart files necessary for an exact restart of the mode
  • A printed log of the run (i.e. stdout output)
    Dummy component models will typically only produce a printed log and possibly a restart file.

    Below we describe in detail:

    Case Naming Conventions

    CSM runs are identified by a unique case name. The case name consists of three parts: a single letter indicating the run type, followed by a three digit run number then a period and a three digit version number. The table below outlines the alphabetic run type abbreviations for the various component model configurations. Run numbers increment slowly to indicate major changes in the component models, while version numbers increment quickly over minor changes with in the current run number. "Active" indicates and active model, while "dummy" indicates and inactive model where "model output" is provided from an existing data files.

    Case Format atm lnd ocn ice
    a000.00 active active dummy dummy
    b000.00 active active active active
    c000.00 dummy dummy active dummy
    d000.00 dummy dummy dummy active
    e000.00 active active active dummy
    f000.00 active active dummy active
    g000.00 dummy dummy active active

    The typical sequence of runs would be:

  • phase 1)   fnnn.mm ~ atm & lnd spinup: CCM3, LSM, CSIM (with specified ice-extent), and docn
  • phase 2)   gnnn.mm ~ ocn & ice spinup: datm, dlnd, CSIM, and NCOM
  • phase 3)   bnnn.mm ~ fully active: CCM3, LSM, CSIM and NCOM

    Mass Storage System (MSS) Naming Conventions

    The MSS is an archival file system at NCAR. Output from CSM runs will be archived on the MSS with pathnames of the form:

      /USER/csm/CASE/atm/... output from atmospheric model
      /USER/csm/CASE/cpl/... output from coupler
      /USER/csm/CASE/ice/... output from ice model
      /USER/csm/CASE/lnd/... output from land model
      /USER/csm/CASE/ocn/... output from ocean model
      /USER/csm/CASE/log/... stdout output from all models

    Where USER is the MSS user name of the person who ran the simulation (typically their login name in all capital letters) and CASE is the case name for the integration.

    The /atm, /cpl, /ice, lnd, and /ocn MSS directories contain the history and restart data files for the respective component models. The specific filenames of the individual history and restart files are generated by the individual component models and may vary.

    The /log directory contains the printed log files output from each component. At the end of each run, the printed log files for each component model are assembled into one compressed tar file and written to the MSS. The /log directory also contains the run scripts used to run the simulation.

    MSS Retention Period and Passwords

    MSS retention periods for production runs will be 3650 days (10 years). MSS write passwords should be the users login name in lower case (e.g. the $LOGNAME environment variable). No read passwords are required. For test runs, retention periods and passwords are left to the discretion of the user.