Research Computing Use Case

Carbon-Aware Research Computing & HPC

Optimize computational research workloads for sustainability without sacrificing scientific output. Schedule simulations, data analysis, and HPC jobs during low-carbon grid hours while meeting grant requirements.

The Research Computing Challenge

Academic and research institutions run compute-intensive workloads—genomic sequencing, climate modeling, molecular simulations, data analysis—often on fixed schedules or queued in traditional job schedulers. These workloads typically have flexible completion windows but lack carbon optimization, missing opportunities to reduce both emissions and costs.

Grant Compliance

Sustainability reporting requirements

Long Runs

Multi-hour to multi-day simulations

Budget Constraints

Limited research computing budgets

Benefits for Research Institutions

Sustainability Reporting

Meet grant requirements and institutional sustainability goals with automated carbon tracking and reduction through intelligent scheduling.

  • Automated carbon accounting per job
  • Export metrics for grant reporting
  • Align with institutional climate goals

Budget Optimization

Reduce electricity costs and maximize compute time within fixed research budgets by scheduling during off-peak, low-cost hours.

  • Lower electricity costs per job
  • More compute with same budget
  • Track cost savings per research group

Perfect for These Research Workloads

Computational Simulations

Long-running simulations in physics, chemistry, climate science, or engineering that can complete within flexible time windows (hours to days).

Example: Molecular dynamics, climate modeling, computational fluid dynamics, finite element analysis

Bioinformatics & Genomics

Genome sequencing, protein folding, phylogenetic analysis, and other compute-intensive bioinformatics pipelines with non-immediate result requirements.

Example: RNA-seq analysis, whole genome sequencing, protein structure prediction

Data Analysis & Processing

Large-scale data processing, statistical analysis, and machine learning model training for research purposes with flexible deadlines.

Example: Research data preprocessing, statistical modeling, exploratory data analysis

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!

3. Track Metrics

Built-in Prometheus metrics automatically track carbon emissions, energy consumption, and cost savings per job—perfect for grant reporting.

Start Sustainable Research Computing Today

Free open-source scheduler for Kubernetes. Perfect for academic institutions and research labs.

Stay Updated on Research Computing Sustainability