Buckets:

glennmatlin's picture
download
raw
1.21 kB
#!/usr/bin/env bash
#SBATCH --job-name=rq4-worker
#SBATCH --partition=ice-cpu
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=1
#SBATCH --mem=8G
#SBATCH --time=01:00:00
#SBATCH --array=0-31
#SBATCH --output=/storage/ice-shared/cs7634/staff/TDA/logs/rq4_array_%A_%a.out
#SBATCH --error=/storage/ice-shared/cs7634/staff/TDA/logs/rq4_array_%A_%a.err
set -euo pipefail
SCRIPT="/storage/ice-shared/cs7634/staff/TDA/code/rq4_bin_characterization.py"
MANIFEST="${MANIFEST:-/storage/ice-shared/cs7634/staff/TDA/soc149/working_sample_manifest.parquet}"
SHARDS_DIR="${SHARDS_DIR:-/storage/ice-shared/cs7634/staff/TDA/trackstar/shards_10k/sample_10000_docs}"
CHUNK_COUNT="${CHUNK_COUNT:-32}"
OUTDIR="${OUTDIR:-/storage/ice-shared/cs7634/staff/TDA/outputs/rq4_workers}"
WORKER_DIR="$OUTDIR/$SLURM_ARRAY_TASK_ID"
mkdir -p "$WORKER_DIR"
echo "RQ4 worker $SLURM_ARRAY_TASK_ID/$CHUNK_COUNT starting at $(date)"
echo "Node: $(hostname)"
python3 "$SCRIPT" \
--mode worker \
--manifest "$MANIFEST" \
--shards-dir "$SHARDS_DIR" \
--chunk-index "$SLURM_ARRAY_TASK_ID" \
--chunk-count "$CHUNK_COUNT" \
--worker-output-dir "$WORKER_DIR"
echo "Worker $SLURM_ARRAY_TASK_ID finished at $(date)"

Xet Storage Details

Size:
1.21 kB
·
Xet hash:
8463b25c3c48b62ea03a6625fcd9e92359c24e65625e8f569c7cafc3aeb8865a

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.