This topic introduces Dify on DMS, a powerful feature integrated into Alibaba Cloud Data Management (DMS). It lets you securely and efficiently build custom AI applications on your databases without writing complex code or migrating data. Examples include AI-powered data Q&A, enterprise knowledge bases, and SQL code generators.
The user owns the container environment of the Dify instance.
Background information
Dify: A popular open-source, visual platform for developing large language model (LLM) applications. It provides a full set of tools to create, orchestrate, and operate AI applications, including prompt engineering, context management, and a retrieval-augmented generation (RAG) engine. For more information about Dify, see the official Dify documentation.
DMS: An enterprise-grade data management and development platform by Alibaba Cloud that provides secure and efficient database management services.
Dify on DMS: A deep integration of Dify with DMS. It embeds the Dify platform directly into the DMS environment, allowing your databases, such as MySQL, PostgreSQL, and SQL Server, to seamlessly and securely serve as knowledge sources for Dify applications. You no longer need to build complex data synchronization pipelines, which enables AI applications to directly interpret your business data.
Why choose Dify on DMS? (Core advantages)
In addition to the features of the community edition of Dify, Dify on DMS provides the following advantages:
Advantage | Description |
Data security | AI applications access your data directly within the DMS environment. Data remains in the database and is not exposed to the public internet, which maximizes the security of your core enterprise data assets. |
Zero data migration | Eliminate the need for extract, transform, and load (ETL) processes or data synchronization. Connect directly to data sources managed by DMS for real-time, native access to business data. This reduces architectural complexity and maintenance costs. |
Low-code development | Dify's visual interface lets business users and developers rapidly build AI applications like assembling blocks, without requiring deep algorithmic expertise. |
Diverse application scenarios | Empower business innovation by easily building practical AI applications, such as natural language-to-SQL converters, AI chatbots, internal knowledge base Q&A, and financial report analysis bots. |
Cost-effective | Explore and deploy generative AI applications at a minimal cost by reusing your existing DMS instances and database resources. |
Version selection guide
Two versions are available to meet different needs. Select a version from the following table based on your requirements.
Feature | Dify on DMS (Community-optimized) | Dify Enterprise Edition |
Base source | Based on the Dify open source version | Official Dify Enterprise Edition |
Core advantages | Optimized by DMS for performance and database compatibility. Production-ready. | Includes all enterprise-grade features and services from the official Dify. Provides deployment and hosting services on Alibaba Cloud. |
Logon method | Supports DMS accounts or an independent Dify account system. | An independent Dify account system. |
Activation method | Ready to use immediately after activation. For more information, see Dify Account Edition User Guide. | Requires a license from the official provider for activation. For more information, see Introduction to Dify on DMS Enterprise Edition. |
Scenarios | For users who want a stable, high-performance, open-source alternative on the cloud. | For users who require full enterprise features and support from the official Dify. |
Billing
The cost of Dify on DMS consists of two parts: DMS compute resource fees and associated application database fees.
Billable item | Dify Community-Optimized Edition | Dify Enterprise Edition |
Dify compute resources | Billed based on Compute Units (CUs) used to run Dify instances. Both subscription and pay-as-you-go billing methods are supported. For specific pricing, see the billing items of DMS compute resources. | |
Dify application databases | You must pay for the databases that Dify requires to run, including the following: Additionally, Dify on DMS supports various engines such as PolarDB for PostgreSQL, AnalyticDB for MySQL, Lindorm, and PolarSearch to offer flexible deployment options. For the billing of specific products, see the documentation for each product. You can also deploy Dify using an existing instance. | |
Dify Enterprise Edition License | Not applicable | For details, contact Alibaba Cloud sales. |
Typical application scenarios
AI-powered data Q&A bot
Scenario: A marketing operations specialist needs to know, "Who was the top salesperson in the Shanghai region last quarter?" without asking a data analyst to write SQL.
Implementation: Create a Q&A bot based on order and user tables. Operations staff can ask questions in natural language, and the AI automatically queries the database and returns the results.
Natural language to SQL generation
Scenario: Junior developers and data analysts need to write complex SQL queries, which is time-consuming and error-prone.
Implementation: Create an SQL generation tool. Input a prompt such as "Query the average spending of all active users in the last 30 days", and the AI automatically generates accurate SQL code.
Private enterprise knowledge base
Scenario: An enterprise needs to build an intelligent Q&A system from product documents, FAQs, and technical manuals stored in a database for employees or customers to use.
Implementation: Connect relevant data tables to Dify as a knowledge base. The AI can then accurately answer user questions based on this content.