1. 31 Oct, 2019 1 commit
  2. 06 Feb, 2019 1 commit
  3. 07 Jul, 2017 1 commit
    • Andrew Heather's avatar
      INT: Integration of Mattijs' collocated parallel IO additions · d8d6030a
      Andrew Heather authored
      Original commit message:
      Parallel IO: New collated file format
      When an OpenFOAM simulation runs in parallel, the data for decomposed fields and
      mesh(es) has historically been stored in multiple files within separate
      directories for each processor.  Processor directories are named 'processorN',
      where N is the processor number.
      This commit introduces an alternative "collated" file format where the data for
      each decomposed field (and mesh) is collated into a single file, which is
      written and read on the master processor.  The files are stored in a single
      directory named 'processors'.
      The new format produces significantly fewer files - one per field, instead of N
      per field.  For large parallel cases, this avoids the restriction on the number
      of open files imposed by the operating system limits.
      The file writing can be threaded allowing the simulation to continue running
      while the data is being written to file.  NFS (Network File System) is not
      needed when using the the collated format and additionally, there is an option
      to run without NFS with the original uncollated approach, known as
      The controls for the file handling are in the OptimisationSwitches of
          //- Parallel IO file handler
          //  uncollated (default), collated or masterUncollated
          fileHandler uncollated;
          //- collated: thread buffer size for queued file writes.
          //  If set to 0 or not sufficient for the file size threading is not used.
          //  Default: 2e9
          maxThreadFileBufferSize 2e9;
          //- masterUncollated: non-blocking buffer size.
          //  If the file exceeds this buffer size scheduled transfer is used.
          //  Default: 2e9
          maxMasterFileBufferSize 2e9;
      When using the collated file handling, memory is allocated for the data in the
      thread.  maxThreadFileBufferSize sets the maximum size of memory in bytes that
      is allocated.  If the data exceeds this size, the write does not use threading.
      When using the masterUncollated file handling, non-blocking MPI communication
      requires a sufficiently large memory buffer on the master node.
      maxMasterFileBufferSize sets the maximum size in bytes of the buffer.  If the
      data exceeds this size, the system uses scheduled communication.
      The installation defaults for the fileHandler choice, maxThreadFileBufferSize
      and maxMasterFileBufferSize (set in etc/controlDict) can be over-ridden within
      the case controlDict file, like other parameters.  Additionally the fileHandler
      can be set by:
      - the "-fileHandler" command line argument;
      - a FOAM_FILEHANDLER environment variable.
      A foamFormatConvert utility allows users to convert files between the collated
      and uncollated formats, e.g.
          mpirun -np 2 foamFormatConvert -parallel -fileHandler uncollated
      An example case demonstrating the file handling methods is provided in:
      The work was undertaken by Mattijs Janssens, in collaboration with Henry Weller.
  4. 25 Jan, 2016 1 commit
    • mattijs's avatar
      ENH: glboal file handling: initial commit · e4240592
      mattijs authored
      Moved file path handling to regIOobject and made it type specific so
      now every object can have its own rules. Examples:
      - faceZones are now processor local (and don't search up anymore)
      - timeStampMaster is now no longer hardcoded inside IOdictionary
        (e.g. uniformDimensionedFields support it as well)
      - the distributedTriSurfaceMesh is properly processor-local; no need
        for fileModificationChecking manipulation.
  5. 03 Feb, 2015 1 commit
    • Henry's avatar
      blockMesh: Add support for multi/sectional grading in a block · 7ec17dfd
      Henry authored
      Consider a block describing a channel with two opposite walls.
      Currently in order to grade the mesh towards the walls and have a
      uniform region in the centre the channel would need to be spit into 3
      blocks.  With the new multi/sectional grading this can be achieved in a
      single block e.g.
          hex (0 1 2 3 4 5 6 7) (20 60 20)
              ((0.2 0.3 4) (0.6 0.4 1) (0.2 0.3 0.25))
      In this example the block is divided uniformly in the x and z -directions
      and split into three grading sections in the y-direction described by
      three triples:  ((0.2 0.3 4) (0.6 0.4 1) (0.2 0.3 0.25)).  Each of the
      grading sections is described by a triple consisting of the fraction of
      the block, the fraction of the divisions and the grading ratio (size of
      first division/size of last division).  Both the fraction of the block
      and the fraction of the divisions are normalized automatically so they
      can be specified scaled in anyway, e.g. as percentages:
          hex (0 1 2 3 4 5 6 7) (20 60 20)
              ((2 3 4) (6 4 1) (2 3 0.25))
      and they need not sum to 1 or 100.
      This is very new functionality and not well tested but backward
      compatibility has been well tested so all existing blockMeshDicts should
      parse correctly.
  6. 14 Aug, 2011 1 commit
  7. 19 Jan, 2011 1 commit
  8. 14 Jan, 2011 1 commit
  9. 07 Jan, 2011 1 commit
  10. 05 Jan, 2011 2 commits
  11. 28 Jul, 2010 1 commit
  12. 29 Mar, 2010 1 commit
  13. 10 Jan, 2009 1 commit
    • Mark Olesen's avatar
      Simplify checking of container (List/HashTable, strings) sizes · 95dcb6de
      Mark Olesen authored
      - can use 'XX.empty()' instead of 'XX.size() == 0', 'XX.size() < 1' or
        'XX.size() <= 0' or for simpler coding.
        It also has the same number of characters as '!XX.size()' and /might/ be
        more readable
      - many size checking had 'XX.size() > 0', 'XX.size() != 0', or 'XX.size() >= 1'
        when a simple 'XX.size()' suffices
  14. 09 Jan, 2009 1 commit
  15. 31 Dec, 2008 1 commit
  16. 25 Jun, 2008 2 commits
  17. 15 Apr, 2008 1 commit