Understanding DPU Consumption
Understand how BEEM measures computational usage through Data Processing Units (DPUs), what operations consume them, and how to optimize your consumption.
BEEM uses Data Processing Units (DPUs) to measure the computational work across your workspace. DPUs provide transparent usage tracking for all data operations.
Check Your Current Usage
- Look at the lower-left corner of the sidebar
- You will see:
- Current DPU consumption vs your plan limit
- Storage used vs capacity (in GB)
- Days remaining in the billing cycle
- Usage data refreshes daily
Understand What Consumes DPUs
Core Operations
- Pipeline triggers: Every time a pipeline runs (manually or on schedule), DPUs are used
- Data ingestion: Syncing data from external sources through connections
- Dataset refreshes: Computing new or updated results in the Warehouse
- Flows: Running reverse ETL to external tools
Infrastructure and Add-ons
- VPN connections to on-premises data sources
- Data Migration Services with automated scanning
- Bastion hosts for secure network access
- Data gateways for firewall-protected or on-premises data
Choose Your Billing Model
- Pay-As-You-Go: Charges based on actual DPU consumption
- Monthly Included DPUs: Fixed monthly credits with overage charges for usage beyond the included amount
Optimize Your Consumption
- Review schedules: Not every dataset needs to refresh daily. Match the schedule to how often the source data actually changes.
- Consolidate loads: Combine multiple smaller data loads into single jobs where possible
- Check downstream triggers: Review if all automatic cascading refreshes are necessary
- Pause unused connections: Pausing connections that are not actively needed reduces consumption
- Evaluate infrastructure: Review whether all VPN tunnels and gateways are still required
For questions about your plan or usage, contact [email protected].
Updated 4 days ago
