First microservice architecture diagram in Microservices - System Design Guide

Problem Statement

When a single application grows too large, it becomes hard to understand, update, and scale. If one part fails, the whole app can crash, causing downtime and poor user experience.

Solution

Breaking the application into small, independent services lets each part work on its own. Each service handles a specific task and communicates with others through simple messages, so failures are isolated and scaling is easier.

Architecture

API Gateway

→User Service

↓

→Database B

↓

→Payment Service

This diagram shows a simple microservice setup where an API Gateway routes requests to independent services like User, Order, and Payment, each with its own database.

Trade-offs

✓ Pros

→

Services can be developed, deployed, and scaled independently.

→

Failures in one service do not crash the entire system.

→

Teams can work on different services simultaneously without conflicts.

✗ Cons

→

Increased complexity in managing multiple services and their communication.

→

Requires infrastructure for service discovery, load balancing, and monitoring.

→

Data consistency can be harder to maintain across services.

Use when the application is growing beyond a single codebase, with multiple teams working on different features, or when independent scaling of components is needed.

Avoid if the application is small, simple, or if the team lacks experience with distributed systems, as overhead may outweigh benefits.

Real World Examples

Netflix

Netflix split its monolithic video streaming platform into microservices to allow independent scaling and faster feature deployment.

Amazon

Amazon uses microservices to separate different business domains like ordering, payment, and inventory, enabling independent development and scaling.

Uber

Uber adopted microservices to handle different services like ride matching, payments, and notifications independently for better reliability.

Alternatives

Monolithic Architecture

All features are built into a single codebase and deployed as one unit.

Use when: Choose when the application is small, simple, or early in development to reduce complexity.

Modular Monolith

The application is a single deployable unit but organized into modules with clear boundaries.

Use when: Choose when you want some separation of concerns without the overhead of distributed systems.

Summary

Microservices break a large application into small, independent services that communicate over the network.

This approach improves scalability, fault isolation, and team autonomy but adds complexity in management and communication.

It is best suited for growing applications with multiple teams and complex domains.

Practice

1. What is the main role of an API Gateway in a microservice architecture?

easy

A. It stores all the data for the microservices.

B. It routes client requests to the correct microservice.

C. It runs the user interface of the application.

D. It replaces all microservices with a single service.

Solution

1. Step 1: Understand the API Gateway function

   The API Gateway acts as a single entry point that directs client requests to the appropriate microservice.

2. Step 2: Compare other options

   Storing data is done by individual services, not the gateway. The UI runs separately, and the gateway does not replace microservices.

3. Final Answer:

   It routes client requests to the correct microservice. -> Option B

4. Quick Check:

   API Gateway = Request Router [OK]

Hint: API Gateway directs requests, it does not store data [OK]

Common Mistakes:

• Thinking API Gateway stores data
• Confusing API Gateway with UI component
• Assuming API Gateway replaces microservices

2. Which of the following correctly shows a microservice owning its own data?

easy

A. Multiple microservices share one database directly.

B. Microservices do not use databases at all.

C. Each microservice has its own separate database.

D. All microservices write to a single shared file.

Solution

1. Step 1: Recall microservice data ownership principle

   Each microservice should own and manage its own database to avoid tight coupling.

2. Step 2: Evaluate options

   Sharing one database or file breaks independence. Not using databases is unrealistic for data needs.

3. Final Answer:

   Each microservice has its own separate database. -> Option C

4. Quick Check:

   Microservice = Own Data Store [OK]

Hint: Microservices keep data separate, no shared DB [OK]

Common Mistakes:

• Assuming all services share one database
• Thinking microservices don't need databases
• Using shared files for data storage

3. Given this simple microservice setup:
Client -> API Gateway -> Service A -> Service B
What happens if Service B is down when Client sends a request?

medium

A. The API Gateway automatically retries Service B until it responds.

B. The API Gateway routes the request to Service B's backup automatically.

C. The client request is handled fully by Service A without contacting Service B.

D. Service A will fail to complete the request and return an error to the client.

Solution

1. Step 1: Trace request flow with Service B down

   The client request goes through API Gateway to Service A, which calls Service B. If Service B is down, Service A cannot complete its task.

2. Step 2: Understand failure impact

   Without Service B responding, Service A returns an error back through the API Gateway to the client.

3. Final Answer:

   Service A will fail to complete the request and return an error to the client. -> Option D

4. Quick Check:

   Down service causes error response [OK]

Hint: Down service causes error, no automatic retry [OK]

Common Mistakes:

• Assuming automatic retries by API Gateway
• Thinking Service A can handle request alone
• Believing API Gateway has backup routing

4. In this microservice diagram, the API Gateway calls Service A and Service B directly. But Service A calls Service B internally and Service B calls Service A internally.
What is the main problem with this design?

medium

A. It creates a circular dependency between services.

B. API Gateway should not call any services directly.

C. Services should share one database instead.

D. Service A should call Service B, not the other way.

Solution

1. Step 1: Identify service call relationships

   API Gateway calls both Service A and Service B, and Service A calls Service B and Service B calls Service A, forming a loop.

2. Step 2: Understand circular dependency issues

   Circular dependencies cause tight coupling and can lead to failures or deadlocks.

3. Final Answer:

   It creates a circular dependency between services. -> Option A

4. Quick Check:

   Circular calls = Bad design [OK]

Hint: Avoid circular calls between microservices [OK]

Common Mistakes:

• Thinking API Gateway shouldn't call services
• Believing shared database fixes call loops
• Assuming call direction must be reversed

5. You want to design a microservice architecture for an online store with these needs:
- User service manages user profiles
- Product service manages product info
- Order service creates orders and needs user and product data

Which architecture diagram best fits these requirements?

hard

A. Client -> API Gateway -> User Service, Product Service, Order Service; Order Service calls User and Product Services internally.

B. Client -> API Gateway -> Order Service only; Order Service manages users and products internally.

C. Client -> API Gateway -> User Service and Product Service only; Order Service is part of API Gateway.

D. Client -> API Gateway -> User Service; Product Service calls Order Service; Order Service calls User Service.

Solution

1. Step 1: Analyze service responsibilities

   User and Product services manage their own data. Order service needs to create orders using data from both.

2. Step 2: Check communication flow

   Order service calling User and Product services internally keeps responsibilities clear and allows data ownership.

3. Step 3: Evaluate options

   Client -> API Gateway -> User Service, Product Service, Order Service; Order Service calls User and Product Services internally. matches this design. Others either merge services incorrectly or create wrong call flows.

4. Final Answer:

   Client -> API Gateway -> User Service, Product Service, Order Service; Order Service calls User and Product Services internally. -> Option A

5. Quick Check:

   Order service calls needed services [OK]

Hint: Order service calls user and product services internally [OK]

Common Mistakes:

• Merging all logic into one service
• Placing order service inside API Gateway
• Incorrect service call directions