Next: Updating Buffers
Up: Customizing CLUSTEREASY
Previous: CLUSTEREASY Variables
As described in section 7.2, each processor stores
n values in the direction and N values
in each other direction. In addition, each processor stores one extra
``buffer'' value of at each end of its grid for calculating
spatial derivatives. To loop over all gridpoints in 3D, for example,
you would write
|
(7.1) |
The macro LOOP automatically does this looping
correctly. If you are calculating the average of a quantity across the
entire lattice you will need to gather the results from all the
processors after the loop is completed. The simplest way to do this is
with the functions MPI_Reduce() or
MPI_Allreduce(). These are explained in any standard
reference on MPI, and you can find examples of them in CLUSTEREASY.
Go to The
LATTICEEASY Home Page
Go to Gary Felder's Home
Page
Send email to Gary Felder at gfelder@email.smith.edu
Send
email to Igor Tkachev at Igor.Tkachev@cern.ch
This
documentation was generated on 2008-01-21