remove SIF option for now. Only use the latest repo container

convert-to-parquet/run-convert-to-parquet.sh

@@ -11,8 +11,7 @@ mem="16G"
 time="02:00:00"
 partition="amd-hdr100"
 outdir=""
-sif=""
-default_sif_image="daskdev/dask:2024.8.0-py3.12"
+sif="gitlab.rc.uab.edu:4567/mdefende/gpfs-policy:latest"
 
 ############################################################
 # Help                                                     #
@@ -20,37 +19,36 @@ default_sif_image="daskdev/dask:2024.8.0-py3.12"
 usage()
 {
 >&2 cat << EOF
-Usage: $0 [ -h ] [ -s | --sif ] [ -o | --outdir ] [ -n | --ntasks ] [ -p | --partition] [ -t | --time ] [ -m | --mem ] <gpfs_logdir>"
+Usage: $0 [ -h ] [ -o | --outdir ] [ -n | --ntasks ] [ -p | --partition] [ -t | --time ] [ -m | --mem ] gpfs_logdir"
 EOF
 exit 1
 }
 
 help()
 {
-# Display Help
-echo "Submits an array job to convert parts of a GPFS log to parquet format"
-echo
-echo "Syntax: $0 [ -h ] [ -s | --sif ] [ -o | --outdir ] [ -n | --ntasks ] [ -p | --partition] [ -t | --time ] [ -m | --mem ] <gpfs_logdir>"
-echo "options:"
-echo "-h|--help     Print this Help."
-echo 
-echo "Required:"
-echo "   gpfs_log_dir           Directory containing GPFS log outputs"
-echo
-echo "Path:"
-echo "   -s|--sif         Path to SIF containing dask for processing"
-echo "   -o|--outdir      Directory to save parquet files to"
-echo 
-echo "sbatch options:"
-echo "   -n|--ntasks      Number of tasks for each array index (default: 1)"
-echo "   -p|--partition   Partition to submit tasks to (default: amd-hdr100)"
-echo "   -t|--time        Max walltime (default: 02:00:00)"
-echo "   -m|--mem         Memory for each task (default: 16G)"
-echo
-exit 1
+>&2 cat << EOF
+Submits an array job to convert parts of a GPFS log to parquet format
+Syntax: $0 [ -h ] [ -o | --outdir ] [ -n | --ntasks ] [ -p | --partition] [ -t | --time ] [ -m | --mem ] gpfs_logdir
+
+options:
+    -h|--help       Print this Help.
+
+Required:
+    gpfs_log_dir    Directory containing GPFS log outputs
+
+Path:
+   -o|--outdir      Directory to save parquet files to
+
+sbatch options:
+   -n|--ntasks      Number of tasks for each array index (default: 1)
+   -p|--partition   Partition to submit tasks to (default: amd-hdr100)
+   -t|--time        Max walltime (default: 02:00:00)
+   -m|--mem         Memory for each task (default: 16G)
+EOF
+exit 0
 }
 
-args=$(getopt -a -o hs:o:n:p:t:m: --long help,sif:,outdir:,ntasks:,partition:,time:,mem: -- "$@")
+args=$(getopt -a -o ho:n:p:t:m: --long help,outdir:,ntasks:,partition:,time:,mem: -- "$@")
 if [[ $? -gt 0 ]]; then
   usage
 fi
@@ -61,7 +59,6 @@ while :
 do
   case $1 in
     -h | --help)      help           ;;
-    -s | --sif)       sif=$2         ; shift 2 ;;
     -o | --outdir)    outdir=$2      ; shift 2 ;;
     -n | --ntasks)    ntasks=$2      ; shift 2 ;;
     -p | --partition) partition=$2   ; shift 2 ;;
@@ -85,25 +82,31 @@ if [[ -z "$gpfs_logdir" ]]; then
     exit 1
 fi
 
-if [[ -z "${sif}" && ! -f "dask.sif" ]]; then
-    echo "No SIF set, downloading ${default_sif_image} as dask.sif"
-    singularity pull dask.sif docker://${default_sif_image}
-    sif="dask.sif"
-elif [[ -f "${sif}" ]]; then
-    echo "Singualrity file does not exist. Please pull the image first"
-    exit 1
+# If outdir not set, set to ${gpfs_logdir}/parquet
+if [[ -z "$outdir" ]]; then
+    outdir="${gpfs_logdir}/parquet"
 fi
 
+singularity pull --force gpfs.sif docker://${sif}
+
 nlogs=$(ls ${gpfs_logdir}/list-* | wc -l)
 
->&2 echo "sif:                 ${sif}"
->&2 echo "output dir:          ${outdir}"
->&2 echo "GPFS logs:           ${gpfs_logdir}"
->&2 echo "ntasks:              ${ntasks}"
->&2 echo "partition:           ${partition}"
->&2 echo "time:                ${time}"
->&2 echo "mem:                 ${mem}"
->&2 echo "singularity command: singularity exec --bind /data ${sif} python3 convert-to-parquet.py -o ${outdir} -f \${log}"
+cmd="singularity exec --bind /data,/scratch gpfs.sif python3 convert-to-parquet.py -o ${outdir} -f \${log}"
+
+>&2 cat << EOF
+--------------------------------------------------------------------------------
+sif:                 ${sif}
+output dir:          ${outdir}
+GPFS logs:           ${gpfs_logdir}
+
+ntasks:              ${ntasks}
+partition:           ${partition}
+time:                ${time}
+mem:                 ${mem}
+
+command:             ${cmd}
+--------------------------------------------------------------------------------
+EOF
 
 mkdir -p out
 mkdir -p err
@@ -126,7 +129,7 @@ mkdir -p err
 
 log=\$(ls ${gpfs_logdir}/list-* | awk "NR==\${SLURM_ARRAY_TASK_ID} { print \$1 }")
 
-singularity exec --bind /data dask.sif python3 convert-to-parquet.py -o ${outdir} -f \${log}
+${cmd}
 EOF
 
 exit 0
\ No newline at end of file