Difference between revisions of "MPI"
From HCL
(New page: You can choose different MPIs. E.g. Open MPI: http://www.open-mpi.org ) |
|||
| (35 intermediate revisions by 4 users not shown) | |||
| Line 1: | Line 1: | ||
| − | + | == Documentation == | |
| + | * http://www.mpi-forum.org/docs/docs.html | ||
| − | + | == Implementations == | |
| + | * [[LAM]] | ||
| + | * [[MPICH]] | ||
| + | * [[OpenMPI]] | ||
| + | * [[MPICH2]] | ||
| + | |||
| + | == Manual installation == | ||
| + | Install in separate subfolder <code>$HOME/SUBDIR</code>, because you may need some MPI implementations (see [[Libraries]]) | ||
| + | |||
| + | == Tips & Tricks == | ||
| + | * For safe consecutive communications create new context, for example: | ||
| + | <source lang="C"> | ||
| + | int communication_operation(MPI_Comm comm) { | ||
| + | MPI_Comm newcomm; | ||
| + | MPI_Comm_dup(comm, &newcomm); | ||
| + | ... // work with newcomm | ||
| + | MPI_Comm_free(&newcomm); | ||
| + | } | ||
| + | </source> | ||
| + | Mind the overhead of <code>MPI_Comm_dup</code> and <code>MPI_Comm_free</code>. | ||
| + | |||
| + | * If you are having trouble with the multi-homed nature of the HCL Cluster, check [http://www.open-mpi.org/faq/?category=tcp#tcp-selection here] | ||
| + | |||
| + | == Debugging == | ||
| + | * Add the following code: | ||
| + | <source lang="C"> | ||
| + | int rank; | ||
| + | MPI_Comm_rank(MPI_COMM_WORLD, &rank); | ||
| + | if (!rank) | ||
| + | getc(stdin); | ||
| + | MPI_Barrier(MPI_COMM_WORLD); | ||
| + | </source> | ||
| + | * Compile your code with <code>-g</code> option | ||
| + | * Run parallel application | ||
| + | * Attach to process(es) from [[GDB]] | ||
| + | ** MPICH-1 runs a background process for each application process: 0, 0b, 1, 1b, ..., therefore, attach to the first ones. | ||
| + | |||
| + | == Profiling == | ||
| + | |||
| + | [http://www.bsc.es/computer-sciences/performance-tools/paraver/general-overview Paraver] by Barcelona Supercomputing Center is a "a flexible performance visualization and analysis tool" | ||
| + | |||
| + | [http://www.bsc.es/computer-sciences/performance-tools/downloads Download here] and [http://www.bsc.es/computer-sciences/performance-tools/documentation tutorials here]. | ||
| + | |||
| + | Use Extrae to create trace files. | ||
| + | |||
| + | Configered and installed extrae on Grid5000 with: | ||
| + | |||
| + | ./configure --prefix=$HOME --with-papi=$HOME --with-mpi=/usr --enable-openmp --with-unwind=$HOME --without-dyninst | ||
| + | make; make install | ||
| + | |||
| + | Create trace.sh (modified from example extrae file): | ||
| + | |||
| + | <source lang="bash">#!/bin/bash | ||
| + | export EXTRAE_HOME=$HOME | ||
| + | export EXTRAE_CONFIG_FILE=$HOME/bin/extrae.xml | ||
| + | export LD_LIBRARY_PATH=${EXTRAE_HOME}/lib:@sub_MPI_HOME@/lib:@sub_PAPI_HOME@/lib:@sub_UNWIND_HOME@/lib:$LD_LIBRARY_PATH | ||
| + | export LD_PRELOAD=${EXTRAE_HOME}/lib/libmpitrace.so | ||
| + | |||
| + | ## Run the desired program | ||
| + | $*</source> | ||
| + | |||
| + | Using the standard extrae.xml supplied with the package. | ||
| + | |||
| + | mpirun -np 3 ~/bin/trace.sh ./executable | ||
| + | |||
| + | Files created: ''TRACE.mpits, TRACExxxxxx.mpit'' | ||
| + | |||
| + | On head node run: | ||
| + | |||
| + | mpi2prv -f TRACE.mpits -e ./executable -o output_tracefile.prv | ||
| + | |||
| + | On local machine open ''output_tracefile.prv'' with paraver | ||
Latest revision as of 23:01, 16 July 2012
Contents
Documentation
Implementations
Manual installation
Install in separate subfolder $HOME/SUBDIR, because you may need some MPI implementations (see Libraries)
Tips & Tricks
- For safe consecutive communications create new context, for example:
int communication_operation(MPI_Comm comm) {
MPI_Comm newcomm;
MPI_Comm_dup(comm, &newcomm);
... // work with newcomm
MPI_Comm_free(&newcomm);
}
Mind the overhead of MPI_Comm_dup and MPI_Comm_free.
- If you are having trouble with the multi-homed nature of the HCL Cluster, check here
Debugging
- Add the following code:
int rank;
MPI_Comm_rank(MPI_COMM_WORLD, &rank);
if (!rank)
getc(stdin);
MPI_Barrier(MPI_COMM_WORLD);
- Compile your code with
-goption - Run parallel application
- Attach to process(es) from GDB
- MPICH-1 runs a background process for each application process: 0, 0b, 1, 1b, ..., therefore, attach to the first ones.
Profiling
Paraver by Barcelona Supercomputing Center is a "a flexible performance visualization and analysis tool"
Download here and tutorials here.
Use Extrae to create trace files.
Configered and installed extrae on Grid5000 with:
./configure --prefix=$HOME --with-papi=$HOME --with-mpi=/usr --enable-openmp --with-unwind=$HOME --without-dyninst make; make install
Create trace.sh (modified from example extrae file):
#!/bin/bash
export EXTRAE_HOME=$HOME
export EXTRAE_CONFIG_FILE=$HOME/bin/extrae.xml
export LD_LIBRARY_PATH=${EXTRAE_HOME}/lib:@sub_MPI_HOME@/lib:@sub_PAPI_HOME@/lib:@sub_UNWIND_HOME@/lib:$LD_LIBRARY_PATH
export LD_PRELOAD=${EXTRAE_HOME}/lib/libmpitrace.so
## Run the desired program
$*
Using the standard extrae.xml supplied with the package.
mpirun -np 3 ~/bin/trace.sh ./executable
Files created: TRACE.mpits, TRACExxxxxx.mpit
On head node run:
mpi2prv -f TRACE.mpits -e ./executable -o output_tracefile.prv
On local machine open output_tracefile.prv with paraver
