Using SCOOP for distributed parallel processing

README2SCOOP.rst

@@ -0,0 +1,21 @@
+******************************************************
+SCOOP enabled RMG-Py
+******************************************************
+
+RMG-Py can be run in parallel (only for the thermochemical parameter 
+estimation part) using SCOOP module.
+More info on SCOOP: http://code.google.com/p/scoop/
+
+Running RMG-Py in parallel:
+
+python -m scoop.__main__  -n 8 $RMGpy/rmddg.py input.py > RMG.sdout.log &
+
+-n 8 specifies that you will have 8 workers. 
+Set it based on the available number of processors.
+For job submission scripts check examples/rmg/scoop.
+
+Installing SCOOP:
+
+You need the development version of SCOOP (tagged with 0.7RC2).
+Download link: http://scoop.googlecode.com/archive/0.7RC2.zip
+

examples/rmg/scoop/input.py

@@ -0,0 +1,159 @@
+# Data sources
+database(
+    thermoLibraries = ['primaryThermoLibrary','DFT_QCI_thermo','GRI-Mech3.0'],
+    reactionLibraries = [('Methylformate',False),('Glarborg/highP',False)],
+    seedMechanisms = ['Glarborg/C2'],
+    kineticsDepositories = ['training'],
+    kineticsFamilies = ['!Intra_Disproportionation'],
+    kineticsEstimator = 'rate rules',
+)
+
+# List of species
+species(
+    label='Mfmt',
+    reactive=True,
+    structure=SMILES("COC=O"),
+)
+species(
+    label='O2',
+    reactive=True,
+    structure=SMILES("[O][O]"),
+)
+species(
+    label='C2H',
+    reactive=True,
+    structure=SMILES("C#[C]"),
+)
+species(
+    label='CH',
+    reactive=True,
+    structure=adjacencyList(
+        """
+        1     C     3 {2,S}
+        2     H     0 {1,S}
+        """),
+)
+species(
+    label='H2O',
+    reactive=True,
+    structure=SMILES("O"),
+)
+species(
+    label='H2',
+    reactive=True,
+    structure=SMILES("[H][H]"),
+)
+species(
+    label='CO',
+    reactive=True,
+    structure=SMILES("[C]=O"),
+)
+species(
+    label='CO2',
+    reactive=True,
+    structure=SMILES("C(=O)=O"),
+)
+species(
+    label='CH4',
+    reactive=True,
+    structure=SMILES("C"),
+)
+species(
+    label='CH3',
+    reactive=True,
+    structure=SMILES("[CH3]"),
+)
+species(
+    label='CH3OH',
+    reactive=True,
+    structure=SMILES("CO"),
+)
+species(
+    label='C2H4',
+    reactive=True,
+    structure=SMILES("C=C"),
+)
+species(
+    label='C2H2',
+    reactive=True,
+    structure=SMILES("C#C"),
+)
+species(
+    label='CH2O',
+    reactive=True,
+    structure=SMILES("C=O"),
+)
+species(
+    label='CH3CHO',
+    reactive=True,
+    structure=SMILES("CC=O"),
+)
+
+
+# Bath gas
+species(
+    label='Ar',
+    reactive=False,
+    structure=InChI("InChI=1S/Ar"),
+)
+
+# Reaction systems
+simpleReactor(
+    temperature=(650,'K'),
+    pressure=(1.0,'bar'),
+    initialMoleFractions={
+        "Mfmt": 0.01,
+        "O2": 0.02,
+        "Ar": 0.08,
+    },
+    terminationTime=(0.5,'s'),
+)
+simpleReactor(
+    temperature=(1350,'K'),
+    pressure=(3.0,'bar'),
+    initialMoleFractions={
+        "Mfmt": 0.01,
+        "O2": 0.02,
+        "Ar": 0.97,
+    },
+    terminationTime=(0.5,'s'),
+)
+simpleReactor(
+    temperature=(1950,'K'),
+    pressure=(10.0,'bar'),
+    initialMoleFractions={
+        "Mfmt": 0.01,
+        "O2": 0.02,
+        "Ar": 0.97,
+    },
+    terminationTime=(0.5,'s'),
+)
+
+simulator(
+    atol=1e-22,
+    rtol=1e-8,
+)
+
+model(
+    toleranceKeepInEdge=0.0,
+    toleranceMoveToCore=0.0005,
+    toleranceInterruptSimulation=1.0,
+    maximumEdgeSpecies=100000
+)
+
+pressureDependence(
+    method='modified strong collision', # 'reservoir state'
+    maximumGrainSize=(1.0,'kcal/mol'),
+    minimumNumberOfGrains=200,
+    temperatures=(290,3500,'K',8),
+    pressures=(0.02,100,'bar',5),
+    interpolation=('Chebyshev', 6, 4),
+)
+
+options(
+    units='si',
+    saveRestartPeriod=None,
+    drawMolecules=False,
+    generatePlots=False,
+    saveConcentrationProfiles=True,
+)

examples/rmg/scoop/lsf.sh

@@ -0,0 +1,59 @@
+#!/bin/sh
+#BSUB -o RMG.out
+#BSUB -J RMGPyScoop
+#BSUB -n 8
+#BSUB -e error_log
+#BSUB -q medium_priority
+
+# This is a job submission file for a LSF queuing system to run
+# the SCOOP-enabled parallel version of RMG-Py across 8 CPUs on
+# a number of different compute nodes on a (potentially heterogeneous) cluster.
+
+source ~/.bash_profile
+
+LAMHOST_FILE=hosts
+
+# start a new host file from scratch
+rm -f $LAMHOST_FILE
+touch $LAMHOST_FILE
+# echo "# LAMMPI host file created by LSF on `date`" >> $LAMHOST_FILE
+# check if we were able to start writing the conf file
+if [ -f $LAMHOST_FILE ]; then
+	:
+else
+	echo "$0: can't create $LAMHOST_FILE"
+	exit 1
+fi
+HOST=""
+NUM_PROC=""
+FLAG=""
+TOTAL_CPUS=0
+for TOKEN in $LSB_MCPU_HOSTS
+do
+	if [ -z "$FLAG" ]; then
+		HOST="$TOKEN"
+		FLAG="0"
+	else
+		NUM_PROC="$TOKEN"
+		TOTAL_CPUS=`expr $TOTAL_CPUS + $NUM_PROC`
+		FLAG="1"
+	fi
+	if [ "$FLAG" = "1" ]; then
+		_x=0
+		while [ $_x -lt $NUM_PROC ]
+		do
+			echo "$HOST" >>$LAMHOST_FILE
+			_x=`expr $_x + 1`
+		done
+		# get ready for the next host
+		FLAG=""
+		HOST=""
+		NUM_PROC=""
+	fi
+done
+# last thing added to LAMHOST_FILE
+#echo "# end of LAMHOST file" >> $LAMHOST_FILE
+echo "Your lamboot hostfile looks like:"
+cat $LAMHOST_FILE
+
+python -m scoop -vv --hostfile $LAMHOST_FILE $RMGpy/rmg.py input.py > RMG.stdout.log

examples/rmg/scoop/prolog.sh

@@ -0,0 +1,2 @@
+#!/bin/bash
+source ~/.bash_profile

examples/rmg/scoop/sge.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+#####################i################################################
+# This is a job submission file for a SGE queuing system to run
+# the SCOOP-enabled parallel version of RMG-Py across 48 CPUs on
+# a single node.
+#
+# Define RMGPy as the path to rmg.py in your ~/.bash_profile
+# NSLOTS is an SGE env. variable for total number of CPUs.
+# prolog.sh is a script used by SCOOP to pass env. variables
+#
+# You can run the jobs on different nodes as well, but it is not
+# recommended since you might have problems with SGE job termination.
+# Type `qconf -spl` to see available parallel environments and modify
+# the last #$ line if you really want to run it on many nodes.
+#####################i################################################
+#$ -S /bin/bash
+#$ -cwd
+#$ -notify
+#$ -o job.log -j y
+#$ -N RMGscoop
+#$ -l normal
+#$ -l h_rt=09:05:00 
+#$ -pe singlenode 48
+source ~/.bash_profile
+python -m scoop.__main__ --tunnel --prolog $RMGpy/examples/rmg/scoop/prolog.sh  -n $NSLOTS $RMGpy/rmg.py input.py > std.out