We may not need to use openmpi with Infiniband if we can get similar performance running psana2 on Ethernet for MPI communications. This connections are needed only for transferring small data (11 GB) for this test from Smd0 to EventBuilders and BigData nodes. Here we show the performance of reading 123 GB on 16 files using 7 drp nodes (113 cores: 1 Smd0/ 12 EventBuilders/ 100 Bigdata cores).
Conclusion:
Using OpenMPI with Infiniband: Rate 39.5 kHz (Total Time: 253 s)
Using MPICH from conda on Ethernet: Rate 39.7 kHz (Total Time: 252 s)
Note 1: below are plots from Grafana showing incoming/outgoing traffics
OpenMPI with Infiniband: the outgoing traffic peaks show Smd0 sending chunks of data to EventBuilders over Infiniband.
MPICH on Ethernet: no noticeable peaks
To run the test:
OpenMPI with Infiniband:
- Clone psana environment then remove mpi4py, mpich, and mpi.
- Build openmpi on drp nodes (drp-tst-dev011 was used for this test). No special flag needed just use --prefix to put the build somewhere. Existing pre-built is located at ~monarin/tmp/4.0.0-rhel7.
- Build mpi4py using this openmpi (see recipe on relmanage/recipe)
cd ~/tmp
git clone https://github.com/mpi4py/mpi4py.git
cd mpi4py
export PATH=/cds/home/m/monarin/tmp/4.0.0-rhel7/bin:$PATH
which mpicc
python setup.py install --single-version-externally-managed --record=record.txt
- Running it with (for example),
~/tmp/4.0.0-rhel7/bin/mpirun --hostfile openmpi_hosts --mca btl_openib_allow_ib 1 run_slac.sh
where ~/tmp/4.0.0-rhel7/ is the --prefix used to build openmpi.
(ps-1.2.2-openmpi) monarin@drp-tst-dev011 (master *) psana2 $ cat openmpi_hosts drp-tst-dev011 slots=1
drp-tst-dev012 slots=12
drp-tst-dev013 slots=20
drp-tst-dev014 slots=20
drp-tst-dev015 slots=20
drp-tst-dev016 slots=20
drp-tst-dev017 slots=20
(ps-1.2.2-openmpi) monarin@drp-tst-dev011 (master *) psana2 $ cat run_slac.sh
#!/bin/bash
export PS_SMD_NODES=12
source $HOME/lcls2/setup_env.sh
conda activate ps-1.2.2-openmpi
python dev_bd.py
MPICH on ethernet
(ps-2.1.2) monarin@drp-tst-dev011 (master *) psana2 $ /reg/g/psdm/sw/conda2/inst/envs/ps-2.1.2/bin/mpirun -f mpich_hosts ./run_slac.sh
(ps-2.1.2) monarin@drp-tst-dev011 (master *) psana2 $ cat mpich_hosts
drp-tst-dev011:1
drp-tst-dev012:12
drp-tst-dev013:20
drp-tst-dev014:20
drp-tst-dev015:20
drp-tst-dev016:20
drp-tst-dev017:20
(ps-1.2.2-openmpi) monarin@drp-tst-dev011 (master *) psana2 $ cat run_slac.sh
#!/bin/bash
export PS_SMD_NODES=12
source $HOME/lcls2/setup_env.sh
python dev_bd.py
Note:
Below is another test to run MPICH on Infiniband using -iface flag. Test test is done under ps-3.1.8.
mpirun -f mpich_hosts -iface ib0 ./run_slac.sh
(ps-3.1.8) monarin@drp-tst-dev011 (master *) psana2 $ cat run_slac.sh
#!/bin/bash
source $HOME/lcls2/setup_env.sh
conda activate ps-3.1.8
python test_send_recv.py
(ps-3.1.8) monarin@drp-tst-dev011 (master *) psana2 $ cat test_send_recv.py
from mpi4py import MPI
comm = MPI.COMM_WORLD
rank = comm.Get_rank()
size = comm.Get_size()
myhost = MPI.Get_processor_name()
import numpy as np
n = 100000
if rank == 0:
data = np.arange(1000000, dtype='i')
for i in range(n):
comm.send(data, dest=1)
comm.send(np.zeros(1000000, dtype='i'), dest=1)
else:
while True:
data = comm.recv(source=0)
print(f'rank{rank} recv {data.shape} {np.sum(data)}')
if np.sum(data) == 0:
break
print(f'rank{rank} on host {myhost} done')
Checking the traffic on IB
Note on Slurm
To submit slurm job, use following two methods
sbatch submit_slac.sh
cat submit_slac.sh
#!/bin/bash
#SBATCH --partition=anagpu
#SBATCH --job-name=psana2-test
#SBATCH --ntasks=4
#SBATCH --ntasks-per-node=4
#SBATCH --output=%j.log
# -u flushes print statements which can otherwise be hidden if mpi hangs
t_start=`date +%s`
srun python ./test_mpi.py
t_end=`date +%s`
echo PSJobCompleted TotalElapsed $((t_end-t_start))
or
srun --partition=anagpu --ntasks=4 --ntasks-per-node=4 python ./test_mpi.py
Overview
Content Tools