Notebook in Data Management (DMS) combines large language models (LLMs) with interactive documents to streamline data delivery and self-service analytics. You can deliver queried data, test data, and data change trends as documents, and use built-in tools to answer data-related questions.
Background information
Data fabric is a data management approach that focuses on rapid, business-oriented data service delivery. According to Gartner, it can reduce data management labor costs by 50% and workload by 70%, accelerating the transformation of data into value. The integration of data fabric and AI enhances data delivery flexibility and simplifies analytics, making data-driven insights accessible to everyone.
Data analytics and applications in DMS
DMS uses data fabric and large language models (LLMs) to build a data management foundation for analytics and applications. The four core features are:
-
Security hosting: Based on database access control best practices at Alibaba Group, DMS provides permission management solutions that help enterprises centrally manage database permissions across multiple clouds. For more information, see Security hosting.
-
DMS Data Copilot: An intelligent data assistant built on Alibaba Cloud LLMs. It combines the data management capabilities of DMS to help developers, O&M engineers, and product staff manage data more efficiently and in a standardized way. For more information, see Data Copilot (New).
-
Notebook: An interactive document that combines code, text, and charts on a single page for efficient data queries and trend visualization.
NoteUsers own the resources on which notebooks run.
-
AI agent: A customizable, publishable AI agent that serves as a unified data service layer, allowing users to query and analyze data through natural language.
Notes
This feature is in public preview and is free of charge during this period. If you have any questions, contact DMS technical support (DingTalk ID: 44962304).
This feature is available only for DMS deployed in the China (Hangzhou), China (Beijing), China (Shanghai), and China (Shenzhen) regions.
Features
|
Intended user |
Before you begin |
DMS solution |
Benefit |
|
Data developers and analysts |
Spend considerable time meeting frequent data query requests from business teams. |
|
Focus on high-value delivery. |
|
Requires proactive manual maintenance due to a lack of valid business metadata. |
Automatically generates business metadata based on LLM inference and business feedback. |
Business metadata is maintained automatically with minimal manual intervention. This improves metadata management efficiency by 50%. |
|
|
Data users, such as product staff, operations engineers, and managers |
Data is hard to find, response times are long, and self-service is not available. |
Data query agents built by developers let users retrieve data through natural language on multiple channels, such as the web and DingTalk, with no code required. |
|