Release May 14, 2026

This release (Release May 14, 2026, v2.2) introduces a series of new features, enhancements, and fixes. Please note that these updates will be rolled out gradually to the following regions. The rollout will be completed within one to two weeks from the release date, depending on your region.

• Domestic Regions

  • Alibaba Cloud (Shanghai)
  • Tencent Cloud (Shanghai/Beijing/Guangzhou)
  • AWS (Beijing)
  • Alibaba Cloud (Hangzhou)

• International Regions

  • Alibaba Cloud (Singapore)
  • AWS (Singapore)

Breaking Changes

• None

AI Capability Updates

• DataAgent: New intelligent data pipeline creation capability, covering end-to-end intelligent support from requirement understanding to execution.

Users simply describe their business needs, and DataAgent automatically generates a modeling plan that conforms to data warehouse layering requirements based on internal or industry-standard specifications. Once the plan is confirmed, a complete data pipeline can be created with a single click, including data integration, data processing and cleansing, and other full-process tasks — from modeling design to pipeline deployment, the entire process is Agent-driven, eliminating the need for manual step-by-step configuration.

In addition, DataAgent automatically outputs a complete task deliverable list and supports detailed visualization capabilities such as DAG diagram display.

New Features

• Data Integration: For offline sync tasks, support for mapping functions and parameters from source data sources of MySQL, PostgreSQL, and SQL Server types to target table fields.
• Data Integration: Offline sync for PostgreSQL data source type now supports syncing Vector type fields.
• Data Integration: Offline sync now supports checkpoint-based resume capability for syncing from Doris source to Lakehouse data source.
• Task Development: Task parameters support encrypted values. During task development, users can set encrypted values for sensitive parameter fields, storing critical information such as passwords, keys, and tokens in ciphertext form. Once encryption is enabled, the specific value of that field will be automatically hidden in task configurations, execution logs, and all run records, eliminating plaintext exposure of sensitive information at the source and comprehensively ensuring data security.

• Task Development: Support for conditional branching. During task development, users can configure conditional branches for tasks based on business logic, enabling differentiated execution path control. When a task runs to a branch node, the system automatically routes to the corresponding processing branch based on preset conditions, without manual intervention. Branch Task

• Task Development: When running combined task node types in temporary mode, users can flexibly select the schema type value from the task scheduling configuration, allowing ad-hoc verification without modifying the formal task configuration.

• Job History: The job history details page has been optimized and improved, including displaying output table information on the plan topology diagram, optimizing the data volume details on connections between stages, and clicking a stage to view more time period and concurrency information.

• Operations Center: Optimized the pause/resume logic for scheduled tasks, allowing users to manually select the specific time state for resuming instances.

Known Limitations

• None

Internal Notice