MaxCompute:Compute pricing (pay-as-you-go)

Specifications

Scaling method

Billing method

Elastically reserved CUs

• Scheduled scaling

• The system scales out CUs at scheduled times based on quota rules and time plans, then scales them in when the period ends. Configure schedules in Manage compute resources - Quotas.

• Scale in resources promptly when no longer needed. You are charged for elastically reserved CUs as long as they remain scaled out, whether or not they are used.

• Pay-as-you-go

• No upfront payment required. Fees follow the formula: User-specified amount of elastic reserved CUs × Actual usage duration. Billing starts when scale-out completes and CUs are allocated to your subscription quota as dedicated resources. Billing stops when scale-in completes.

• Metering data is pushed hourly and whenever CU count changes. The system aggregates total usage in CU-hours per level-1 quota to generate bills.

Auto-scaling CUs

• Automatic scaling

• Once enabled in a quota rule, the system detects workload changes and scales resources up to the limit you set. You only configure the maximum CU count — no manual scale-out or release needed. Configure in Manage compute resources - Quotas.

• You are charged only for the resources you actually use.

• Pay-as-you-go

• No prepayment required. The system monitors usage at per-second granularity and charges based on auto-scaling CU usage × actual usage duration during the auto-scaling period.

• Bills are generated hourly by aggregating total CU-hour usage per level-1 quota.

Type

Public cloud pricing

Finance Cloud pricing

Notes

Elastically reserved CUs

0.0488 USD/CU/hour

CNY 0.5985 per CU per hour

If daily usage exceeds 16 hours, subscription CUs are more cost-efficient.

Auto-scaling CUs

0.0558 USD/CU/hour

CNY 0.684 per CU per hour

Best practices for auto-scaling