Master Architecture Analysis: System, Decomposition, Process & Object Thinking

This article, from an introductory architecture practice course, introduces architecture analysis methods: system engineering thinking, system decomposition thinking, process‑oriented analysis, object‑oriented analysis, and key points. It also compares monolithic (ALL IN ONE) and distributed system architectures from different perspectives.

1. Architecture Analysis Methods – Outline

System engineering thinking

System decomposition thinking

Process‑oriented thinking

Object‑oriented thinking

How architects should analyze

1.1 System Engineering Thinking

System engineering: Treat the system and its related matters as a whole.

1. System perspective: consider all stakeholders and requirements.

2. Process perspective: cover the entire lifecycle from requirements to release, trial operation, and eventual decommission.

3. Team perspective: meet team needs, structure architecture layers, and select technologies.

1.2 System Decomposition Thinking

System decomposition: divide the system into manageable granularity for easier analysis and problem solving.

1. Split large system into secondary systems.

2. Break complex problems into simple ones.

3. Separate generic and specific problems.

4. Separate stable and unstable parts.

1.3 Process‑Oriented Analysis

Process‑oriented: use workflow and modular thinking, top‑down analysis.

Identify an entry point, then analyze each step and influencing factors.

Key principles:

Top‑down / bottom‑up

Whole first, then parts (or large before small)

Modular / workflow

System decomposition is also a process analysis method

1.4 Object‑Oriented Analysis

Object‑oriented: everything is an object; program = object + interaction; system = domain model + interaction.

Key concepts:

Classes, objects, relationships

Inheritance, encapsulation, polymorphism

Class diagrams, component diagrams, deployment diagrams, package diagrams

Use case, activity, sequence, collaboration, state diagrams

1.5 How Architects Should Analyze

Integrate system, decomposition, process, and object thinking, and any other applicable mindsets for architecture.

2. System Structural Composition

Monolithic system structure (ALL IN ONE)

Distributed system structure (large complex system)

Basic elements of system composition (logical view)

2.1 Monolithic System Structure (ALL IN ONE)

System description

A B/S‑based Customer Relationship Management system with features:

Simple customer information management

Customer classification

View purchase records

Customer follow‑up

Uses Oracle database for data storage.

Analysis results (partial view)

2.2 Distributed System Structure (Large Complex System)

System description

An e‑commerce system with features:

Users can purchase goods online

Shop owners can apply to open stores, publish and manage products

Users and shop owners can view their orders

Platform provides online support to close transaction loops

Integrates logistics platform for real‑time delivery

Platform offers store decoration functionality

Uses MySQL cluster for data storage, caches hot data, accelerates static data via CDN, and adopts distributed architecture with service decomposition.

Analysis results (partial view)

Deployment perspective

Each system deployed in independent clusters

Reverse proxy or multi‑level load balancing

CDN operator data‑center deployment and synchronization

MySQL database cluster

Distributed cache cluster, MQ cluster

2.3 Basic Elements of System Composition (Logical View)

From a logical perspective, a system can be divided into subsystems, modules, components, interfaces, classes, objects, databases, tables, records, etc.