SLURM Cron Job Setup for Dashboard Sync
Resource Usage Analysis
Per Sync Job: - Memory: 512MB (very light) - CPU: 1 core - Time: Maximum 5 minutes (typically 30-60 seconds) - Network: Minimal git operations only
Total Daily Load: - If run every 15 minutes: 96 jobs/day × 1-2 minutes = ~2-3 hours total CPU time - If run every 30 minutes: 48 jobs/day × 1-2 minutes = ~1-1.5 hours total CPU time
Recommended Crontab Entry
Add to your cluster crontab (crontab -e):
# Dashboard sync every 15 minutes (recommended for active development)
*/15 * * * * cd ~/research/mosquito-alert-model-monitor && sbatch scripts/slurm_dashboard_sync.sh
# OR: Dashboard sync every 30 minutes (recommended for production)
*/30 * * * * cd ~/research/mosquito-alert-model-monitor && sbatch scripts/slurm_dashboard_sync.shLoad Impact Assessment
MINIMAL LOAD - This is extremely lightweight: - Uses same resources as a simple git operation - Runs for only 1-2 minutes typically - No computational processing - Only file I/O and network operations
Comparison: - Your data prep jobs: ~hours of runtime, GBs of memory - Dashboard sync: ~1 minute runtime, 512MB memory
Recommendation: Start with 15-minute intervals. If you want to reduce load further, change to 30-minute intervals.
Monitoring
Check sync job status:
# View recent sync jobs
squeue -u $USER | grep dashboard_sync
# Check sync logs
tail -50 ~/research/mosquito-alert-model-monitor/logs/dashboard_sync.logManual Sync
If needed, run sync manually:
cd ~/research/mosquito-alert-model-monitor
sbatch scripts/slurm_dashboard_sync.sh