All Products
Search
Document Center

Data Management:Notebook

Last Updated:Dec 16, 2024

Notebook of Data Management (DMS) works with large language models (LLMs) to empower business developers, data developers, analysts, and data operators and improve the efficiency of data delivery and self-service data analysis. You can use Notebook to deliver information such as the data to be queried, test data, and data trends as documents. Data Workstation provides tools that can answer data-related questions based on the delivered documents.

Background information

Data fabric is an innovative data management approach that focuses on the rapid delivery of business-oriented data services and aims at more efficient and convenient data usage. According to Gartner, this technology can reduce the labor cost for data management by 50%, reduce the workload by 70%, and accelerate the transformation of data values. With the development of AI technology, the integration of data fabrics and AI enhances the flexibility of data delivery and lowers the difficulty of data analysis. This makes data analysis easier for everyone and ushers in a new era in which everyone can participate in data analysis.

Use DMS for data analysis and application

DMS uses data fabrics and LLMs to construct a data management base, which facilitates data analysis and application. The database management base provides the following core features:

  • Secure hosting: the best practice of DMS for database access control in Alibaba Group. DMS provides enterprises with a collection of database permission management solutions and helps enterprises manage database permissions across clouds in a centralized manner. This ensures that users can safely use data. For more information, see Security hosting.

  • DMS Data Copilot: an intelligent data assistant built by DMS based on the LLMs of Alibaba Cloud. DMS Data Copilot integrates with the proficient data management and data application capabilities of DMS to help users such as developers, O&M engineers, and product staff use and manage data in a more efficient and standard manner.

  • Notebook: an interactive tool that allows users to combine code, text, and charts on one page. This allows users to efficiently query data and predict data trends in a visual manner.

    Note

    The resources on which notebooks run belong to users.

  • AI agent: an agent that can be customized and published. An AI agent can be used as a unified data service layer for external users. Users can interact with an AI agent by using natural languages. This allows users to query and analyze data by using natural languages.

Usage notes

This feature is supported only in the China (Hangzhou) and Singapore regions.

Features

Intended user

Before Notebook is enabled

DMS solution

Benefit

Data developers and analysts

Data developers and analysts spend considerable time to meet frequent data query requirements of the business party.

  • Provides Data Copilot that supports capabilities such as natural language to SQL (NL2SQL), SQL completion, and SQL error correction. This helps developers easily develop SQL statements at a high efficiency.

  • Provides an agent configuration and debugging platform. This allows developers to create data query chatbots with a few clicks. The agents are built on out-of-the-box custom LLMs and support uninterrupted services to meet over 80% repetitive data query requirements.

Data developers and analysts can focus on high-value delivery.

Manual proactive O&M is required due to lack of valid business metadata.

Automatically obtains business metadata based on LLM inference and business feedback.

Business metadata is automatically maintained, assisted by manpower. The metadata management efficiency is improved by 50%.

Data users, including product staff, operations engineers, and managers

It is difficult to query data, the response is slow, and self-help services are not supported.

Uses the agents created by developers to provide data query services over multiple channels, such as web and DingTalk. Users can interact with the agents by using natural languages. They need to only submit questions to obtain the results. The entire process involves zero coding.

  • Self-service data usage with low skill requirements is supported. This allows users to focus on business decision-making.

  • The agents provide 24/7 services. This eliminates the need for queuing or waiting.

References