Changeset 622 for palm/trunk/SOURCE/advec_particles.f90

Timestamp:

Dec 10, 2010 8:08:13 AM (13 years ago)

Author:

raasch

Message:

New:
---

Optional barriers included in order to speed up collective operations
MPI_ALLTOALL and MPI_ALLREDUCE. This feature is controlled with new initial
parameter collective_wait. Default is .FALSE, but .TRUE. on SGI-type
systems. (advec_particles, advec_s_bc, buoyancy, check_for_restart,
cpu_statistics, data_output_2d, data_output_ptseries, flow_statistics,
global_min_max, inflow_turbulence, init_3d_model, init_particles, init_pegrid,
init_slope, parin, pres, poismg, set_particle_attributes, timestep,
read_var_list, user_statistics, write_compressed, write_var_list)

Adjustments for Kyushu Univ. (lcrte, ibmku). Concerning hybrid
(MPI/openMP) runs, the number of openMP threads per MPI tasks can now
be given as an argument to mrun-option -O. (mbuild, mrun, subjob)

Changed:

---

Initialization of the module command changed for SGI-ICE/lcsgi (mbuild, subjob)

Errors:

---

File:

• palm/trunk/SOURCE/advec_particles.f90 (modified) (4 diffs)

palm/trunk/SOURCE/advec_particles.f90

@@ -4 +4 @@
 ! Current revisions:
 ! -----------------
+! optional barriers included in order to speed up collective operations
 ! TEST: PRINT statements on unit 9 (commented out)
 !
@@ -792 +793 @@
 !
 !--       Compute total sum from local sums
+          IF ( collective_wait )  CALL MPI_BARRIER( comm2d, ierr )
           CALL MPI_ALLREDUCE( sums_l(nzb,1,0), sums(nzb,1), nzt+2-nzb, &
                               MPI_REAL, MPI_SUM, comm2d, ierr )
+          IF ( collective_wait )  CALL MPI_BARRIER( comm2d, ierr )
           CALL MPI_ALLREDUCE( sums_l(nzb,2,0), sums(nzb,2), nzt+2-nzb, &
                               MPI_REAL, MPI_SUM, comm2d, ierr )
@@ -830 +833 @@
 !
 !--       Compute total sum from local sums
+          IF ( collective_wait )  CALL MPI_BARRIER( comm2d, ierr )
           CALL MPI_ALLREDUCE( sums_l(nzb,8,0), sums(nzb,8), nzt+2-nzb, &
                               MPI_REAL, MPI_SUM, comm2d, ierr )
+          IF ( collective_wait )  CALL MPI_BARRIER( comm2d, ierr )
           CALL MPI_ALLREDUCE( sums_l(nzb,30,0), sums(nzb,30), nzt+2-nzb, &
                               MPI_REAL, MPI_SUM, comm2d, ierr )
+          IF ( collective_wait )  CALL MPI_BARRIER( comm2d, ierr )
           CALL MPI_ALLREDUCE( sums_l(nzb,31,0), sums(nzb,31), nzt+2-nzb, &
                               MPI_REAL, MPI_SUM, comm2d, ierr )
+          IF ( collective_wait )  CALL MPI_BARRIER( comm2d, ierr )
           CALL MPI_ALLREDUCE( sums_l(nzb,32,0), sums(nzb,32), nzt+2-nzb, &
                               MPI_REAL, MPI_SUM, comm2d, ierr )
@@ -1948 +1955 @@
 !--    and set the switch corespondingly
 #if defined( __parallel )
+       IF ( collective_wait )  CALL MPI_BARRIER( comm2d, ierr )
        CALL MPI_ALLREDUCE( dt_3d_reached_l, dt_3d_reached, 1, MPI_LOGICAL, &
                            MPI_LAND, comm2d, ierr )