DataService Studio is a flexible, lightweight, and secure serverless platform in DataWorks for building and managing data APIs. It acts as a bridge between databases and data applications—letting you generate REST APIs from data tables, register existing APIs for unified management, and publish them through API Gateway to individuals, teams, and enterprises, without managing infrastructure.
With its serverless architecture, you only need to focus on the query logic of your API. DataService Studio handles compute resources, supports scalability, and requires no operations and maintenance (O&M).
DataService Studio requires Google Chrome version 69 or later on a personal computer (PC).
Use cases
-
Expose data warehouse tables as APIs: Connect MaxCompute, Hologres, or other data sources and generate query APIs in minutes—no coding required with the Codeless UI, or write custom SQL with the Code Editor.
-
Centralize API management: Register APIs already running in your systems into DataService Studio for unified publishing, access control, and metering through API Gateway.
-
Automate data delivery: Use Data Push to schedule SQL-based queries and push results—as rich text or tables—to DingTalk groups, Lark groups, WeCom groups, email, or Microsoft Teams.
Key concepts
| Concept | Definition |
|---|---|
| API | An interface that lets applications query data from your databases and data tables. DataService Studio APIs support read operations only. |
| Function | A filter that processes request or response parameters of an API. The same function type must be used for both the pre-filter and the post-filter. |
| Data Push | A feature that runs SQL statements against data sources such as MySQL, PostgreSQL, Hologres, MaxCompute, and ClickHouse, then pushes the results to messaging channels on a schedule. |
| Resource group for DataService Studio | An exclusive resource group dedicated to a single enterprise. Use it when you need guaranteed QPS and SLA for API calls. |
Billing
API calls in DataService Studio consume compute resources from a resource group. Choose a billing mode based on your QPS requirements and cost preferences.
| Mode | Resource group type | How you're charged | When to use |
|---|---|---|---|
| Pay-as-you-go | Serverless | Number of current CUs × Duration of CU allocation. Billing starts when you allocate Occupied CUs for the DataService Studio purpose. |
Variable or unpredictable traffic |
| Subscription | Serverless | Based on the number of Minimum CUs that you use for the DataService Studio purpose. | Stable, predictable workloads |
| Exclusive resource group | Exclusive (dedicated to one enterprise) | Based on resource group specifications. | High QPS requirements or strict SLA guarantees |
For CU allocation steps, see Allocate CU quotas to tasks. For exclusive resource group pricing, see Billing of exclusive resource groups for DataService Studio.
Public resource groups for DataService Studio are no longer recommended. See Announcement on the deprecation of public resource groups for DataService Studio in DataWorks for details.
API development workflow
The standard workflow takes an API from data source to a callable endpoint in seven steps.
| Step | Task | Description |
|---|---|---|
| 1 | Create a data source and configure network connectivity | Add your database or data warehouse as a DataWorks data source. DataService Studio uses this connection to read table schemas when you configure parameters, and to execute queries when the API is called. If the connectivity test fails, see Network connectivity and whitelist configuration. |
| 2 | Create a business process | Create a business process to organize your APIs and related files such as functions. DataService Studio uses business processes as the development unit. |
| 3 | Create an API | Generate an API or register an existing one. See Choose your API creation method below. |
| 4 | Create a function (filter) | Optional. Use Function Compute to process request or response parameters before or after the API executes. |
| 5 | Test the API | Test the API before publishing. The API cannot be published until it passes testing. |
| 6 | Publish the API | Publish to API Gateway. After publishing, other services can call the API. You can also list it on the Alibaba Cloud API Marketplace to monetize your data. The API Marketplace covers seven categories: finance, artificial intelligence, e-commerce, transportation and geography, lifestyle services, enterprise management, and public affairs. |
| 7 | Call the API | Use an application identity to call the API. Grant or request permissions using Authorize an API. |
Choose your API creation method
At step 3, pick the approach that fits your situation:
| Method | Best for | Guide |
|---|---|---|
| Codeless UI | Generating APIs directly from data tables, no SQL required | Generate an API in the codeless UI |
| Code Editor | Complex queries, table joins, and aggregate functions | Generate an API in the code editor |
| Register an existing API | Registering an API you've already built for unified publishing and management | Register an API |
Where to start
| If you... | Start here |
|---|---|
| Are setting up DataService Studio for the first time | Begin at step 1 |
| Already have a running API you want to manage centrally | Go directly to Register an API |
| Want to push query results to a messaging channel | See Data Push |
Data Push
Data Push lets you write SQL to query your data sources and automatically deliver the results to DingTalk groups, Lark groups, WeCom groups, email, or Microsoft Teams on a schedule. Customize the content format (rich text or tables), push frequency, and recipients based on your business needs. See Data Push.