Pay-as-you-go is a post-payment billing model. You are billed based on the actual compute resources your workspace consumes, with no advance purchase required.
EMR Serverless Spark uses pay-as-you-go billing. During peak periods, resources may be preempted and timely resource availability cannot be fully guaranteed.
When to use pay-as-you-go
Pay-as-you-go works best when:
Your workload varies frequently
Your resource needs are temporary or bursty
Billing rule
Each billing cycle is 1 hour. The fee for a cycle is:
Compute Unit (CU) usage within 1 hour × hourly unit price in your regionAll resources consumed in the cycle are converted to CU-hours, then multiplied by the hourly unit price.
Discount: If your workspace's cumulative CPU-to-memory ratio within an hour falls between 1:2 (inclusive) and 1:3 (exclusive), a 20% discount applies when converting resources to CU.
The formula applies to a single EMR Serverless Spark workspace. If you have multiple workspaces, your bill is the sum of fees across all workspaces.
Billing starts when the container runtime begins — before Spark Driver and Executor initialization. The CU-hours shown in the Spark UI (based on Driver and Executor runtimes) will therefore differ from your actual billed CU-hours. Check your final bill for accurate charges.
Example calculation
Consider the following Spark job configuration:
spark.driver.cores 1
spark.driver.memory 1g
spark.driver.memoryOverhead 1g
spark.executor.cores 1
spark.executor.memory 1g
spark.executor.memoryOverhead 1g
spark.executor.instances 2This job uses 3 cores and 6 GB of memory per minute.
Over one hour, it consumes 180 core-minutes and 360 GB-minutes.
Apply the CU conversion formula:
max(180/1, 360/4) = max(180, 90) = 180 CU-minutesThe CPU-to-memory ratio is 1:2, so the 20% discount applies:
180 CU-minutes × 0.8 = 144 CU-minutes = 2.4 CU-hoursTotal cost for one hour: 2.4 CU-hours × hourly unit price in your region.
For unit prices by region, see Regional unit prices.
Billing cycle
Fees are calculated hourly on the hour (UTC+8). After each calculation, a new billing cycle begins, and the system generates a bill and deducts the corresponding amount from your account.
Billing records may lag behind actual usage.
Regional unit prices
Prices below are for reference. Actual prices are shown on the product purchase page.
Chinese mainland
| Region | Unit price (USD/CU/hour) |
|---|---|
| China (Beijing) | 0.050302 |
| China (Shanghai) | 0.050302 |
| China (Hangzhou) | 0.050302 |
| China (Shenzhen) | 0.050302 |
| China (Hong Kong) | 0.074048 |
Other regions
| Region | Unit price (USD/CU/hour) |
|---|---|
| Indonesia (Jakarta) | 0.067106 |
| Germany (Frankfurt) | 0.064792 |
| Singapore | 0.067106 |
| US (Virginia) | 0.053801 |
| US (Silicon Valley) | 0.053801 |
| Japan (Tokyo) | 0.069420 |
| Mexico | 0.051220 |
Mainland China
Region Name
Unit price (USD/CU/hour)
China (Beijing)
China (Shanghai)
China (Hangzhou)
China (Shenzhen)
0.050302
China (Hong Kong)
0.074048
Other regions
Region
Unit price (USD/CU/hour)
Indonesia (Jakarta)
0.067106
Germany (Frankfurt)
0.064792
Singapore
0.067106
US (Virginia)
0.053801
US (Silicon Valley)
0.053801
Japan (Tokyo)
0.069420
Mexico
0.051220