DataWorks introduces the Serverless Resource Group, which unifies the core features of the legacy exclusive resource groups for scheduling, Data Integration, and data service. A single Serverless Resource Group now handles all core operations, including Data Synchronization, periodic scheduling tasks, and API service, greatly simplifying resource management. This resource group offers two billing models:
Subscription: Provides stable, predictable, and dedicated compute resources, making it ideal for production environments.
Pay-as-you-go: Provides on-demand compute resources with Auto Scaling, offering both flexibility and cost-effectiveness.
Using a Serverless Resource Group incurs a Scheduling Instance Fee when Node Tasks are published to the Production Environment for periodic scheduling.
Billing scenarios
The fees for a DataWorks Serverless resource group consist of a Resource Usage Fee and a Task Scheduling Fee.
Resource Usage Fee: You are charged for the Compute Units (CUs) that specific DataWorks tasks consume in a Serverless resource group. This fee is based on total CU consumption.
A Compute Unit (CU) is defined as
1 CU = 1 virtual CPU (vCPU) + 4 GiB memory.Task Scheduling Fee: This fee applies to periodically scheduled tasks that run in the production environment. These tasks incur only a Task Scheduling Fee and not a Resource Usage Fee. This fee is based on the number of successfully run instances, excluding dry runs.
A Serverless resource group supports a maximum of 200 concurrent instances. This limit meets the maximum concurrency requirements of all legacy resource group specifications, so you do not need to consider CU specifications for scheduling concurrency.
The following diagram and table describe the relationship between supported task types and the fees they incur in a Serverless resource group.
Task type | Description | Fee type |
Data Integration | Runs a Data Synchronization task, such as an offline synchronization task, in Data Integration or Data Studio. | Resource Usage Fee |
Data Compute |
Important For information about Data Compute tasks, see Appendix 1: Task types and CU consumption. | |
Data Service | Calls an API generated in DataService Studio. | |
Personal Development Environment | Uses a Personal development environment to debug tasks. | |
Large Model Service | Deploys and uses a large model service. | |
Task Scheduling | A periodic scheduling task runs in the production environment. | Task Scheduling Fee |
Notes
Pay-as-you-go Serverless Resource Groups can experience resource contention during peak hours, so immediate availability is not guaranteed.
You can convert a pay-as-you-go Serverless Resource Group to a subscription-based one, but you cannot convert it back.
When you first activate DataWorks, it creates a pay-as-you-go Serverless Resource Group by default. You pay only for the resources you use. For billing details, see Pay-as-you-go resource group billing.
Performance specifications
Billing for a Serverless Resource Group is based on the Compute Units (CUs) consumed. A CU is defined as 1 CU = 1 vCPU + 4 GiB memory. Plan your resource group specifications based on your specific development scenarios and task types.
The following recommended specifications are general guidelines. Adjust the resources based on your business requirements to ensure efficient and stable task execution.
Data integration
Batch synchronization
Concurrency configuration | Recommended specifications | Minimum specifications |
< 4 | 0.5 CU | 0.5 CU |
>= 4 |
|
Real-time synchronization
Task type | Recommended specifications | Minimum specifications | |
MySQL real-time synchronization | 1 database | 2 CU | A minimum of 1 CU is required to run a single real-time synchronization task. |
2 to 5 databases | 2 CU | ||
6 or more databases | 2 CU | ||
Kafka real-time synchronization | 1 CU | ||
Other single-table real-time tasks | 1 CU | ||
Entire-database real-time synchronization | - | A minimum of 2 CU is required to run a single entire-database synchronization task. | |
Data compute
Each Data Compute task has a default CU value. For more information, see Appendix 1: Task types and CU consumption.
DataService Studio
Maximum QPS | Minimum specifications | SLA |
500 | 4 CU | 99.95% |
1,000 | 8 CU | |
2,000 | 16 CU |
Personal development environment
Resource quotas for personal development environments range from 2 to 100 CUs for CPU-based tasks and from 21 to 60 CUs for GPU-based tasks. Estimate your needs by task type:
Lightweight tasks (such as simple SQL queries or Python script debugging): Use a lower resource quota, such as 2 CUs.
Moderately complex tasks (such as data processing or Notebook analysis): Use a medium resource quota, such as 4 CUs.
Deep learning tasks (such as TensorFlow or PyTorch model training): Use a GPU-based resource type. Select the appropriate GPU memory and number of CUs based on the model size.
Large model service
Calculate the required CUs based on the GPU memory.
Deploying
0.6B,1.7B,4B, and8Bmodels requires a minimum of24 GBof GPU memory.Deploying a
14Bmodel requires a minimum of48 GBof GPU memory.Deploying a
32Bmodel requires a minimum of96 GBof GPU memory.
Task scheduling
A Serverless Resource Group supports a maximum of 200 concurrent instances. This limit is independent of the CU specifications. The default concurrency is 50 instances. You can set the maximum number of concurrent instances to 200 on the resource group details page.
Billing
Serverless Resource Groups are available in two billing models: Subscription (prepaid) and Pay-as-you-go (postpaid).
Subscription Serverless Resource Group: You pre-pay for a set amount of CUs for a fixed period. This fee covers all resource costs for tasks in DataWorks, such as Data Synchronization, Data Compute, and debugging or calling DataService Studio APIs. You incur no additional resource fees for these tasks.
Pay-as-you-go Serverless Resource Group: You pay for resources after using them, based on the total CUs consumed. Certain tasks, such as Batch Synchronization Tasks, DataService Studio tasks, and Data Development tasks, incur resource fees.
The following table compares the features of the two billing models.
Item | Pay-as-you-go | Subscription |
Total available CUs | Billing is based on actual usage. | You get the number of CUs specified at purchase. |
Scale-out, scale-in, and renewal | Not applicable | Supported |
Quota Management | Controls the maximum CU usage for various scenarios. Supported for Data Compute, Data Integration, and DataService Studio. | |
Concurrent Task Instance Limit | Supported. You can run a maximum of 200 Task Instances concurrently. | |
Number of bound Virtual Private Clouds (VPCs) |
| Depends on the number of CUs you purchase.
|
Pricing
Subscription resource group billing
The cost is calculated using the following formula: Cost = Monthly unit price × Number of months × Number of CUs purchased per month.
The Subscription model requires a minimum monthly purchase of 2 CUs. Although there is no upper limit on purchases, all transactions are subject to inventory. If inventory is insufficient, a notification appears on the purchase page.
If your purchased resources are insufficient, you can scale up at any time. For more information, see Use serverless resource groups.
To see the minimum resources required for different task types in a Serverless Resource Group, see Performance metrics.
Region | Monthly Unit Price (/USD/Month/CU) |
China (Shanghai), China (Hangzhou), China (Beijing), China (Shenzhen) | 37.1517 |
UK (London) | 51.01286 |
US (Virginia) | 53.92014 |
Malaysia (Kuala Lumpur) | 63.36534 |
China (Hong Kong), Singapore, Germany (Frankfurt), Indonesia (Jakarta) | 67.61327 |
US (Silicon Valley) | 72.74794 |
Japan (Tokyo) | 77.45584 |
South Korea (Seoul) | 52.47334281 |
UAE (Dubai) | 81.09255 |
Pay-as-you-go resource group billing
The cost is calculated using the following formula: Cost = CU-hour × CU unit price. Bills are generated hourly.
In resource quota management, if you allocate 1 CU to DataService Studio, billing for that CU continues even if the service is not used. To stop the charges, you must adjust the CU allocation for DataService Studio to 0.
Region | Unit Price (USD/CU-hour) | Example |
China (Shanghai), China (Hangzhou), China (Beijing), China (Shenzhen) | 0.077399 | Example: If a Data Synchronization task in the Shanghai region is configured with 2 CUs, runs for 0.5 hours, and the unit price of a CU is USD 0.077399/CU-hour, then the CU-hours consumed and the cost for this task are as follows:
|
UK (London) | 0.106277 | |
US (Virginia) | 0.112334 | |
Malaysia (Kuala Lumpur) | 0.132011 | |
Germany (Frankfurt), Indonesia (Jakarta), China (Hong Kong), Singapore | 0.140861 | |
US (Silicon Valley) | 0.151558 | |
Japan (Tokyo) | 0.161366 | |
South Korea (Seoul) | 0.10931946 | |
UAE (Dubai) | 0.168943 |
Billing details
The billable items and billing codes for Serverless Resource Groups in the Billing & Cost Management console are as follows:
Pay-as-you-go: The billable item is
General Resource Group CU*H (Pay-as-you-go), and the billing code isexresource_cu_hour_post.Subscription: The billable item is
General Exclusive Resource Group (Subscription and Pay-as-you-go), and the billing code iscu_number.
For details, see View bill details.
Expiration and renewal
You can renew your Subscription Serverless Resource Group before it expires. Failure to renew results in service suspension and the eventual release of the resource group. For details, see Expiration and renewal.
Next steps
You can use a resource group for tasks such as Data Integration, Data Development, and Data Service. For instructions on how to purchase a resource group, attach it to a workspace, and configure its network settings, see Use serverless resource groups.
More information
Appendix 1: Task types and CU consumption
Tasks in DataWorks are either Compute Tasks, which consume CUs, or Scheduling Tasks, which do not.
Identifying task types
Go to the node editing page in Data Studio. In the right-side navigation pane, navigate to to identify the task type.
Compute Task: In the Scheduling Policies section, you must specify the CUs required to run the task.
Scenario 1: You can customize the number of CUs.
Scenario 2: You can only use the default number of CUs.
Scheduling Task: In the Scheduling Policies section, you only need to select a scheduling Resource Group. CU configuration is not required.
CU configuration for compute tasks
Running a Compute Task with a Serverless Resource Group consumes CUs. This section explains the difference between Default CUs and Running CUs:
Default CU: The recommended number of CUs that the platform allocates each time a task runs, based on the task type. Using fewer CUs than the default may reduce task efficiency.
Running CUs: The actual number of CUs configured to run the task. By default, this is set to the Default CU value, which you can adjust as needed. Follow these configuration principles:
The minimum configuration is 0.25 CU, with increments of 0.25 CU. If the message The CU Quota for the current Resource Group is insufficient appears, you can adjust the CU Quota for the Compute Task.
To prevent under- or over-provisioning, configure this parameter based on the Default CU value and the CU Quota for the Compute Task. For more information, see Assign CU Quotas to tasks.
You can only adjust the running CUs for some tasks. For example:
You cannot adjust the running CUs for a Hologres SQL task. It can only be set to 0.25 (the default CU).
The default running CU for a PyODPS 2 task is 0.5, which you can adjust as needed (for example, to 0.4 or 0.6).
Node type | Node name | Default CU | Customizable |
Notebook | 0.5 | Yes | |
MaxCompute | 0.5 | Yes | |
0.5 | Yes | ||
0.5 | Yes | ||
0.25 | Yes | ||
0.25 | Yes | ||
Hologres | 0.25 | - | |
0.25 | - | ||
0.25 | Yes | ||
0.25 | Yes | ||
EMR | 0.25 | - | |
0.25 | - | ||
0.25 | Yes | ||
0.25 | - | ||
0.25 | Yes | ||
0.5 | Yes | ||
0.5 | Yes | ||
0.5 | Yes | ||
0.25 | - | ||
0.25 | - | ||
Serverless Spark | 0.25 | - | |
0.25 | - | ||
0.25 | - | ||
Serverless StarRocks | 0.25 | - | |
Large Language Model (LLM) | 0.5 | - | |
ADB | 0.25 | Yes | |
0.25 | Yes | ||
0.25 | - | ||
0.25 | - | ||
CDH | 0.25 | - | |
0.5 | Yes | ||
0.25 | - | ||
0.25 | - | ||
0.25 | - | ||
0.25 | - | ||
Lindorm | 0.25 | - | |
0.25 | - | ||
ClickHouse | 0.25 | - | |
Data Quality | 0.25 | - | |
0.5 | Yes | ||
General | 0.25 | Yes | |
0.25 | Yes | ||
0.25 | - | ||
0.5 | Yes | ||
0.25 | Yes | ||
0.25 | Yes | ||
0.25 | - | ||
0.25 | - | ||
0.25 | - | ||
MySQL node | 0.25 | - | |
SQL Server node | |||
Oracle node | |||
PostgreSQL node | |||
StarRocks node | |||
DRDS node | |||
PolarDB MySQL node | |||
PolarDB PostgreSQL node | |||
Doris node | |||
MariaDB node | |||
SelectDB node | |||
Redshift node | |||
SAP HANA node | |||
Vertica node | |||
DM node | |||
KingbaseES node | |||
OceanBase node | |||
DB2 node | |||
GBase 8a node | |||
Algorithm | 0.25 | - |
Configuration for scheduling tasks
Scheduling Tasks do not consume CUs from the Serverless Resource Group.
Node type | Node name |
Data Integration | |
MaxCompute | |
Flink | |
General | |
Algorithm |
Appendix 2: Billing modes for task execution
When you run a node task in DataWorks, the associated compute fees are not always billed by DataWorks. The billing depends on the underlying compute engine or resource where the task executes. There are three possible scenarios:
When a task is published to the production environment for periodic scheduling, a task scheduling fee is always incurred.
Execution mode | Example node | Compute resource provider | Fee composition |
Mode 1: The Compute Task runs in a Serverless Resource Group. | PyODPS, Shell, Data Integration, Data Quality | Serverless Resource Group | Serverless Resource Group fees only |
Mode 2: The Compute Task runs on a Third-party Engine via a Serverless Resource Group. | EMR Hive, Hologres SQL | Serverless Resource Group + Third-party Engine | Serverless Resource Group fees + Third-party Engine fees |
Mode 3: The Scheduling Task runs on a Third-party Engine via Scheduling Center. | MaxCompute SQL, Flink SQL | Third-party Engine | Third-party Engine fees only |
Appendix 3: Module fee breakdown
When you use a Serverless Resource Group with the following modules, the following fees apply:
Data Integration: When you perform data synchronization, Data Integration tasks run in the Data Integration, Data Studio, and Scheduling Center modules. This consumes resources from the Serverless Resource Group and incurs Data Integration fees. Periodic synchronization tasks also incur task instance scheduling fees.
Data Studio: When you use Data Studio for task development, Compute and Scheduling Tasks run in the Data Studio, Data Quality, and Scheduling Center modules. This consumes resources from the Serverless Resource Group and incurs data compute fees and task instance scheduling fees. Using a personal development environment incurs additional personal development environment fees. Using a Large Language Model (LLM) service or node also incurs LLM service fees.
Data Analysis: When you use Data Analysis for SQL query analysis or to download query results, data Compute Tasks run in the Data Analysis module. This consumes resources from the Serverless Resource Group and incurs data compute fees. Using the Data Insight feature also incurs task instance scheduling fees.
DataService Studio: When you use DataService Studio to generate an API service from a data source and allocate CUs via resource quota management, this consumes resources from the Serverless Resource Group and incurs Data Service fees. Using the Data Push feature also incurs task instance scheduling fees.