DataWorks Data Development (Data Studio) is undergoing a major architectural upgrade to support Lakehouse architecture, unified batch and stream processing, and native AI integration. The upgrade affects only the Data Studio user interface (UI) and user experience (UX) — published production jobs are not disrupted.
If you encounter any issues during the upgrade, contact the DataWorks Data Development Upgrade Support Group.
Background
DataWorks was originally built 16 years ago. While it served core batch processing needs well, evolving technology and business requirements have exposed architectural limits. This upgrade addresses three areas:
Architectural evolution
Lakehouse architecture support: The new architecture natively supports a smooth transition from traditional data warehouses to a Lakehouse architecture, with unified metadata management and data processing.
Unified batch and stream processing: A unified development paradigm and metadata view seamlessly integrates offline batch processing and real-time stream processing.
Native AI integration: AI capabilities — machine learning, deep learning, and natural language processing — are integrated into the full data development lifecycle.
Technical debt
The legacy monolithic architecture had limited support for cloud-native elasticity and stream processing.
Tightly coupled modules resulted in long iteration cycles, high maintenance costs, and difficulty scaling for large concurrent workloads.
User experience
The new architecture supports tens of thousands of active users processing exabyte-scale data with higher stability, throughput, and resource isolation.
Intelligent features — code assistance, performance diagnostics, and one-click publishing — bring the development experience closer to a local integrated development environment (IDE).
Scope of the upgrade
| Module | Impact |
|---|---|
| Data Development (Data Studio) | UI and UX only |
| Operation Center | Not affected |
| Deployment Center | Not affected |
Before you upgrade
Upgrade eligibility
Only users with Workspace Administrator permissions can see and perform the upgrade.
Impact on existing services
| Item | Impact |
|---|---|
| Published production jobs | Zero impact. All online jobs continue to run without disruption. |
| Core data development features | Fully preserved and enhanced. Node types, function support, and core editing and execution are retained and optimized. |
Asset migration
The following asset types support one-click migration to the new interface:
Existing nodes (such as MaxCompute SQL and Shell)
User-Defined Functions (UDFs)
Resource files (
.jar,.py,.txt)Components
Ad hoc queries
Key constraints — read before starting
The upgrade cannot be rolled back after it completes. Review all constraints below before proceeding. If you need assistance, join the DataWorks Data Development Upgrade Support Group.
Legacy Data Studio OpenAPI users: Contact on-duty engineers in the Upgrade Support Group before starting the upgrade.
Cross-workspace publishing is blocked: After the upgrade, you cannot publish tasks between a workspace running the legacy version and an upgraded workspace.
No edits during the upgrade: While the upgrade is running, you cannot add or modify content in either the legacy or new version of Data Studio for the current workspace. This applies to both UI and OpenAPI operations.
Upgrade timing: The upgrade takes time. Perform it during off-peak development hours.
Workspace identity is unchanged: The workspace name and ID remain unchanged after the upgrade.
Legacy version becomes read-only: After the upgrade, the legacy version of Data Studio is read-only. Code edited in the new version does not sync back to the legacy version.
Ad hoc query file visibility: After the upgrade, migrated ad hoc query files from the legacy version are only visible in your personal directory for files you own.
Upgrade procedure
Step 1: Access the upgrade entry point
In the top navigation bar of the Data Studio main interface, find the blue Upgrade to New Version button next to the workspace selector. Only users with Workspace Administrator permissions can see this button.
Step 2: Review upgrade considerations
After clicking Upgrade to New Version, a page with upgrade information appears. Review all considerations on this page carefully before proceeding.
Step 3: Start the upgrade
When the status shows "Upgrade has not started for the current workspace", click Start Upgrade.
The system displays the estimated upgrade time and the number of objects to be migrated.
Click Confirm to begin.
Monitor upgrade progress in real time.
Click Upgrade Details to view a real-time log of the process.
Click Refresh to check the latest status.
After the upgrade completes, click Go to Data Studio to start using the new version.
Feature mapping
The new Data Studio reorganizes and optimizes all features from the legacy version. Use the table below to find where each legacy feature is now located, then refer to the detailed sections for screenshots.
| Legacy feature | New location | Notes |
|---|---|---|
| Solutions | Focus Mode | Enter from any file directory |
| Business Flow (fixed hierarchy) | Custom directories + tags | Create directories as needed; use tags to replicate Business Flow organization |
| Business Flow Dashboard | Directory Dashboard | The "View" capability is available for any directory |
| Business Flow (workflow orchestration) | Scheduled Workflow | In the Data Development module |
| Manual business flows | Triggered Workflow | Under Project Directory > Workflow > Triggered Workflow; supports event-driven logic |
| Node development (engine-bound only) | Project Directory | All node types visible; write code before binding an engine |
| Run / Run with parameters | Single Run button + Debug Configuration panel | See Run/run with parameters |
| Engine Instance | Data Source + Compute Resource | Split into two distinct concepts |
| Running Resource Group | Resource Group | In the Debug Configuration panel |
| Custom Parameters | Parameters | In the Debug Configuration panel |
| Smoke test (post-submission only) | Standard Mode: after publish to dev env; Basic Mode: after publish to prod env | |
| Code Review (toolbar buttons) | Production Checker step during publishing; list in left directory tree | |
| Submit / Submit and allow others to edit | Publish | Standard Mode: can package and publish to Deployment Center |
| Table development | Data Directory | Create tables via visual form, code, or AI-assisted table creation |
| Table management | Data Directory | |
| Public tables | Data Directory | |
| Resource development | Resource Management | |
| Function development | Resource Management | |
| Function list | Ask Copilot | Not yet available as a dedicated module. A full function management module is planned for a future release |
| Ad hoc queries (all users in workspace) | Ad hoc queries (current user, all workspaces in region) | For code debugging only; to publish as a production job, submit to the project directory and configure scheduling |
| Component management | Component Management (new UI) | |
| Manual tasks | Manual Tasks (new UI) | |
| Operation checks | Operation Checks (new UI) | |
| Run history | Run History (new UI) | |
| Smoke test records | Smoke Test Records (new UI) | |
| Compute resources | Compute Resources (new UI) | |
| Settings | Theme switch + Other settings + Explore more settings | |
| Recycle bin | Recycle Bin (new UI) |
The function list does not have a direct equivalent in the current release. Ask Copilot about function usage. A full function management module — including management and creation of official engine functions and UDFs — is planned for a future release.
Detailed feature changes
Data development
Solutions
Legacy version: A separate "Solutions" menu organized and managed related Business Flows.
New version: This feature is now Focus Mode. Enter Focus Mode from any file directory for a more immersive and flexible development experience.
Business flows
Legacy version: Used a fixed directory hierarchy and supported Business Flow Dashboards.
New version: Create directories as needed and tag them to replicate legacy Business Flow directory organization.
New version: Business Flow Dashboards are now Directory Dashboards. The "View" capability is available for any directory.
New version: For workflow orchestration, create a Scheduled Workflow in the Data Development module.
Node development
Legacy version: Displayed only available node types for bound engines. Node types supported by DataWorks but not yet bound to an engine were not visible, and you could not write code for them.
New version: Node development is integrated into the Project Directory. All node types are visible, and you can write code before binding an engine. The process for creating new nodes is still being optimized to improve the creation and coding experience.
Run/run with parameters
Legacy version: Separate Run and Run with Parameters buttons. Clicking Run with Parameters opened a dialog where you selected a Running Resource Group and Custom Parameters.
New version:
Run and Run with Parameters are consolidated into a single Run button.
The legacy Engine Instance is split into Data Source and Compute Resource, depending on the node type:
Data Source: Used for metadata suggestions in the intelligent code editor.
Compute Resource: Determines the compute resources used for code debugging.
The legacy Running Resource Group is now Resource Group in the Debug Configuration panel.
The legacy Custom Parameters are now Parameters in the Debug Configuration panel.
The Debug Configuration panel is always visible on the right side of the node. Each code run uses the Compute Resource, Resource Group, and Parameters from this panel.
Smoke test/view smoke test records
Legacy version: A Smoke Test could only be started after a successful submission.
New version:
Standard Mode: Start a Smoke Test after successfully publishing the task to the development environment (equivalent to submission in the legacy version).
Basic Mode: Start a Smoke Test after successfully publishing the task to the production environment (equivalent to submission in the legacy version).
Code review/view code review records
Legacy version: Code Review buttons were located directly on the node toolbar.
New version: Initiate a Code Review from the Production Checker step during the publishing process. View the Code Review list in the directory tree on the left.
Submit/submit and allow others to edit
Legacy version:
New version: "Submit" is now Publish. Users with the necessary permissions can publish directly. In Standard Mode, you can also package tasks and publish them to the Deployment Center.
Table development
Legacy version: Tables were created in the Data Development module.
New version: This feature is now the Data Directory management module. Create tables using a visual form, code, or AI-assisted table creation.
Resource development
Legacy version: Resources were created in the Data Development module.
New version: This feature is now part of the Resource Management module.
Function development
Legacy version: Functions were created in the Data Development module.
New version: This feature is now part of the Resource Management module.
Component management
Legacy version:
New version:
Manual tasks
Legacy version:
New version:
Manual business flows
Legacy version:
New version:
For more comprehensive manual workflow capabilities, use Event-driven Workflows. Find this under Project Directory > Workflow > Triggered Workflow.
Ad hoc queries
Legacy version: Displayed ad hoc query files for all users in the current workspace.
New version:
Displays ad hoc query files for the current user across all workspaces in the current region.
This module is for code debugging only. To publish a query as a production job, submit it to the project directory, configure its scheduling settings, and then publish it.
Table management
Legacy version:
New version: This feature is now the Data Directory management module. Create tables using a visual form, code, or AI-assisted table creation.
Public tables
Legacy version:
New version: This feature is now the Data Directory management module.
Function list
Legacy version:
New version: A dedicated function management module is not available in this release. Ask Copilot directly about how to use functions.
A dedicated function management module — including management and creation of official engine functions and UDFs — is planned for a future release.
Operation checks
Legacy version:
New version:
Run history
Legacy version:
New version:
Smoke test records
Legacy version:
New version:
Compute resources
Legacy version:
New version:
Settings
Legacy version:
New version: Theme switch.
New version: Other settings.
New version: Explore more settings.
Recycle bin
Legacy version:
New version:
Upgrade support
| Support type | Details |
|---|---|
| Online help | Detailed guidance is available throughout the upgrade process |
| Technical consultation | Contact the DataWorks Data Development Upgrade Support Group via DingTalk; on-site service available if needed |
| Emergency response | A rapid response mechanism is in place for upgrade-related issues |
| Product training | Training on new features is available upon request |
| Rollback | If you encounter issues during the upgrade, contact the Upgrade Support Group. The team can roll back to the state before the upgrade began. |