ice runner

Author: sbryngelson

.github/workflows/ice/submit.sh

@@ -0,0 +1,60 @@
+#!/bin/bash
+
+set -e
+
+usage() {
+    echo "Usage: $0 [script.sh] [cpu|gpu]"
+}
+
+if [ ! -z "$1" ]; then
+    sbatch_script_contents=`cat $1`
+else
+    usage
+    exit 1
+fi
+
+sbatch_cpu_opts="\
+#SBATCH --ntasks-per-node=20       # Number of cores per node required
+"
+
+sbatch_gpu_opts="\
+#SBATCH -G V100:2\
+"
+
+if [ "$2" == "cpu" ]; then
+    sbatch_device_opts="$sbatch_cpu_opts"
+elif [ "$2" == "gpu" ]; then
+    sbatch_device_opts="$sbatch_gpu_opts"
+else
+    usage
+    exit 1
+fi
+
+job_slug="`basename "$1" | sed 's/\.sh$//' | sed 's/[^a-zA-Z0-9]/-/g'`-$2"
+
+sbatch <<EOT
+#!/bin/bash
+#SBATCH -Jshb-$job_slug            # Job name
+#SBATCH --account=gts-sbryngelson3 # charge account
+#SBATCH -N1                        # Number of nodes required
+$sbatch_device_opts
+#SBATCH -t 02:00:00                # Duration of the job (Ex: 15 mins)
+#SBATCH -q embers                  # QOS Name
+#SBATCH -o$job_slug.out            # Combined output and error messages file
+#SBATCH -W                         # Do not exit until the submitted job terminates.
+
+set -e
+set -x
+
+cd "\$SLURM_SUBMIT_DIR"
+echo "Running in $(pwd):"
+
+job_slug="$job_slug"
+job_device="$2"
+
+. ./mfc.sh load -c p -m $2
+
+$sbatch_script_contents
+
+EOT
+

.github/workflows/ice/test.sh

@@ -0,0 +1,19 @@
+#!/bin/bash
+
+build_opts=""
+if [ "$job_device" == "gpu" ]; then
+    build_opts="--gpu"
+fi
+
+./mfc.sh build -j 8 $build_opts
+
+n_test_threads=8
+
+if [ "$job_device" == "gpu" ]; then
+    gpu_count=$(nvidia-smi -L | wc -l)        # number of GPUs on node
+    gpu_ids=$(seq -s ' ' 0 $(($gpu_count-1))) # 0,1,2,...,gpu_count-1
+    device_opts="-g $gpu_ids"
+    n_test_threads=`expr $gpu_count \* 2`
+fi
+
+./mfc.sh test --max-attempts 3 -a -j $n_test_threads $device_opts -- -c phoenix

.github/workflows/test.yml

@@ -103,7 +103,7 @@ jobs:
     strategy:
       matrix:
         device: ['cpu', 'gpu']
-        lbl: ['gt', 'frontier']
+        lbl: ['gt', 'ice', 'frontier']
         exclude:
           - device:   cpu
             lbl: frontier
@@ -121,6 +121,10 @@ jobs:
         if:   matrix.lbl == 'gt'
         run:  bash .github/workflows/phoenix/submit.sh .github/workflows/phoenix/test.sh ${{ matrix.device }}
 
+      - name: Build & Test
+        if:   matrix.lbl == 'ice'
+        run:  bash .github/workflows/ice/submit.sh .github/workflows/phoenix/test.sh ${{ matrix.device }}
+
       - name: Build
         if:   matrix.lbl == 'frontier'
         run:  bash .github/workflows/frontier/build.sh