Carbon-Aware Batch Processing & Data Pipelines
Schedule ETL jobs, data processing, and batch workloads during low-carbon, off-peak hours to reduce both emissions and electricity costs.
The Opportunity in Batch Workloads
Batch processing jobs—ETL pipelines, data aggregations, report generation, backups—often run on fixed schedules (midnight, hourly, daily) regardless of grid conditions or electricity pricing. These deferrable workloads are perfect candidates for carbon and cost optimization through intelligent scheduling.
Overnight data transformations
Cron-based processing tasks
Time-of-use rate optimization
Dual Optimization: Carbon + Cost
Carbon-Aware Scheduling
Schedule batch jobs to run when grid carbon intensity is lowest, automatically shifting workloads to cleaner energy windows.
- Real-time carbon intensity data
- Configurable thresholds per job
- Maximum delay guarantees (SLA protection)
Price-Aware Scheduling
Optimize for time-of-use (TOU) electricity pricing by scheduling during off-peak hours, reducing infrastructure costs alongside carbon emissions.
- Custom TOU pricing schedules
- Combine carbon + cost optimization
- Prometheus metrics for ROI tracking
Perfect For These Workloads
ETL & Data Transformation
Overnight data pipelines that process logs, aggregate metrics, or transform data for analytics. These jobs often have flexible completion windows (e.g., "must complete by 6 AM").
Example: Daily data warehouse refresh, log aggregation, reporting pipelines
Scheduled Maintenance & Backups
Database backups, index rebuilds, cache warming, and other maintenance tasks that run on cron schedules but don't need precise timing.
Example: Database backups, index optimization, cache pre-population
Periodic Report Generation
Weekly/monthly reports, analytics dashboards, or data exports that can be generated within a time window rather than at an exact moment.
Example: Weekly business reports, analytics exports, compliance reports
Quick Setup
1. Install (2 minutes)
helm repo add compute-gardener https://elevated-systems.github.io/compute-gardener-scheduler helm install compute-gardener-scheduler compute-gardener/compute-gardener-scheduler \ --set carbonAware.electricityMap.apiKey=YOUR_API_KEY
2. Use It (add one line)
spec: schedulerName: compute-gardener-scheduler # That's it!
Optimize Your Batch Workloads Today
Free open-source scheduler for Kubernetes. Start reducing carbon and costs in minutes.