portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-06-16 06:47:06 +01:00
Code Releases Activity

Parallel IO worked on. I'm puzzled because I already thought I shook this out on MacOS + OpenMPI and then

Browse Source 
turned up problems on the BlueWaters Cray.

Gets 75MB/s from home filesystem on parallel configuration read. Need to make the RNG IO parallel,
and also to look at aggregating bigger writes for the parallel write.
Not sure what the home filesystem is.
This commit is contained in:
Peter Boyle
2016-02-21 08:03:21 -06:00
parent 40f2db9bc0
commit 6aeaf6f568
6 changed files with 160 additions and 80 deletions
Download Patch File Download Diff File Expand all files Collapse all files

27
lib/communicator/Communicator_mpi.cc
View File

@ -117,21 +117,22 @@ void CartesianCommunicator::SendToRecvFrom(void *xmit,
SendToRecvFromBegin(reqs,xmit,dest,recv,from,bytes);
SendToRecvFromComplete(reqs);
}
void CartesianCommunicator::RecvFrom(void *recv,
int from,
int bytes)
void CartesianCommunicator::SendRecvPacket(void *xmit,
void *recv,
int sender,
int receiver,
int bytes)
{
MPI_Status stat;
int ierr=MPI_Recv(recv, bytes, MPI_CHAR,from,from,communicator,&stat);
assert(ierr==0);
}
void CartesianCommunicator::SendTo(void *xmit,
int dest,
int bytes)
{
int rank = _processor; // used for tag; must know who it comes from
int ierr = MPI_Send(xmit, bytes, MPI_CHAR,dest,_processor,communicator);
assert(ierr==0);
assert(sender != receiver);
int tag = sender;
if ( _processor == sender ) {
MPI_Send(xmit, bytes, MPI_CHAR,receiver,tag,communicator);
}
if ( _processor == receiver ) {
MPI_Recv(recv, bytes, MPI_CHAR,sender,tag,communicator,&stat);
}
}
// Basic Halo comms primitive