Administration

Cost Management

Monitor, forecast, and optimize your NATIS platform costs using budgets, tags, and cost attribution.

5 min read · Updated May 2025

NATIS bills based on Databricks Units (DBU) consumed by compute clusters. Each cluster node type has a DBU/hour rate that varies by instance size, runtime type (standard vs. ML), and workload type (all-purpose vs. jobs). Costs are tracked in real time in the Cost Dashboard.

Cost Optimization Best Practices

  • Use Job Clusters instead of All-Purpose Clusters for scheduled pipelines (30-50% cheaper)
  • Enable Spot/Preemptible instances for non-critical batch workloads (60-80% savings)
  • Set auto-termination to 30 minutes or less for interactive clusters
  • Use SQL Warehouses (Serverless) for BI queries — cost per query, no idle waste
  • Use OPTIMIZE + VACUUM regularly to reduce storage costs from small file proliferation
  • Apply cluster policies to prevent users from creating oversized clusters
  • Tag all resources with cost center, team, and project for accurate chargebacks

Setting Budget Alerts

Use the NATIS Cost Attribution Tag system to tag pipelines, clusters, and warehouses with team and project codes. This enables accurate chargebacks in multi-team environments without needing separate workspaces.

  • 1. Navigate to Admin Console → Cost Management → Budgets → New Budget.
  • 2. Define the scope: workspace, team group, or individual user.
  • 3. Set the budget period (monthly) and the threshold amount (in USD or DBU).
  • 4. Configure alert thresholds: notify at 50%, 75%, 90%, and 100% of budget.
  • 5. Choose notification recipients and channels.
  • 6. Optionally enable Auto-Stop: automatically terminate new clusters when budget is reached.

Was this page helpful?

Thanks for your feedback!