About this Dify Component Architecture Diagram
This diagram shows dify component architecture diagram in a clearer structure, so the main layers or modules are easier to explain.
Frontend Layer
The Frontend Layer section groups the components that belong to this part of the architecture. In this diagram, it includes Web UI, Mobile App, Desktop Client, API Interface, which makes the boundary of the layer easier to explain when presenting how the system is organized.
- Frontend Layer
- Web UI
- Mobile App
- Desktop Client
- API Interface
- MCP Server
Workflow Engine
The Workflow Engine section marks one visible part of the architecture. In this diagram, it includes Agent Node, LLM Node, Tool Node, Knowledge, so the section reads as a specific functional block rather than a generic label.
- Workflow Engine
- Agent Node
- LLM Node
- Tool Node
- Knowledge
- Code Node
- HTTP Node
- Drag & Drop Visual Builder
Agent and RAG Systems
The Agent and RAG Systems section marks one visible part of the architecture. In this diagram, it includes Agent System, Agent Strategies, Reasoning Engine, Task Planning, so the section reads as a specific functional block rather than a generic label.
- Agent System
- Agent Strategies
- Reasoning Engine
- Task Planning
- Tool Management
- RAG (Retrieval-Augmented Generation)
- ETL
- Embedding
- Vector Index
- Retrieval
Plugin Runtime and Integrations
The Plugin Runtime and Integrations section marks one visible part of the architecture. In this diagram, it includes Plugin System, Model Plugins, Tool Plugins, Extensions, so the section reads as a specific functional block rather than a generic label.
- Plugin System
- Model Plugins
- Tool Plugins
- Extensions
- Bundles
- Custom Tools
- 3rd Party APIs
- Plugin Runtime Engine & Sandbox Environment
Model Runtime and Infrastructure
The Model Runtime and Infrastructure section marks one visible part of the architecture. In this diagram, it includes Model Runtime Layer, OpenAI, Azure OpenAI, Anthropic, so the section reads as a specific functional block rather than a generic label.
- Model Runtime Layer
- OpenAI
- Azure OpenAI
- Anthropic
- Local LLMs
- PostgreSQL
- Redis
- Vector DB
- Object Storage
- Monitoring
- Logging
FAQs about this Template
-
How do IT teams visualize Azure architecture?
IT teams usually visualize Azure architecture with a layered diagram that separates identity, application, data, and infrastructure services. This makes it easier to review service dependencies, deployment boundaries, and operational flow across sections such as Frontend Layer, Workflow Engine, and Agent and RAG Systems, especially when Azure resources need to support both scalability and governance.
-
Can AI generate Azure architecture diagrams automatically?
Yes, AI can generate an initial Azure architecture diagram, but it should not replace technical review. AI can suggest resource groupings and service flow, while engineers still need to confirm the real Azure services, networking setup, security controls, deployment logic, and support assumptions before using the diagram for planning, delivery, or stakeholder communication.
-
What is the difference between Azure architecture and cloud architecture?
Azure architecture is a cloud architecture pattern built specifically around Microsoft Azure services, while cloud architecture is a broader category that also covers AWS, Google Cloud, and hybrid environments. Azure diagrams usually focus more directly on Azure-native resources, service relationships, deployment paths, security controls, and platform-specific integration choices.
-
What should an Azure architecture diagram include?
An Azure architecture diagram should include the main services, user entry points, data handling flow, and the core network or security boundaries. It should also show how application services, storage, identity, monitoring, external integrations, and access controls connect, so the system can be reviewed for design clarity, maintainability, and deployment readiness.
-
Which diagram type is best for Azure solution planning?
A high-level architecture diagram is usually the best choice for Azure solution planning because it shows how major services and boundaries fit together before implementation starts. Teams often pair it with deployment, data flow, or network views later when they need more detail for delivery planning, troubleshooting, compliance review, infrastructure operations, or role alignment.
(AI & Web)