Identifying service boundaries in Microservices - System Design Exercise

Design: Microservices Service Boundary Identification

Focus on identifying and defining service boundaries and their interactions. Out of scope: detailed implementation of each microservice, infrastructure setup, and deployment pipelines.

Functional Requirements

FR1: Decompose a monolithic application into multiple microservices

FR2: Each microservice should own a distinct business capability

FR3: Services should have clear, minimal dependencies on each other

FR4: Support independent deployment and scaling of services

FR5: Ensure data consistency within each service boundary

Non-Functional Requirements

NFR1: System should handle 10,000 concurrent users

NFR2: API response latency p99 under 300ms

NFR3: Availability target of 99.9% uptime

NFR4: Services must communicate asynchronously where possible

NFR5: Data duplication across services should be minimized

Think Before You Design

Questions to Ask

❓ Question 1

❓ Question 2

❓ Question 3

❓ Question 4

❓ Question 5

Key Components

Domain-driven design concepts (bounded contexts)

API gateways or service mesh for communication

Event-driven messaging systems

Databases owned per service

Service registry and discovery

Design Patterns

Bounded Context pattern

Database per service pattern

Event sourcing and CQRS

API Gateway pattern

Saga pattern for distributed transactions

Reference Architecture

                +---------------------+
                |     API Gateway      |
                +----------+----------+
                           |
      +--------------------+--------------------+
      |                    |                    |
+-----v-----+        +-----v-----+        +-----v-----+
| User      |        | Order     |        | Inventory |
| Service   |        | Service   |        | Service   |
+-----------+        +-----------+        +-----------+
      |                    |                    |
+-----v-----+        +-----v-----+        +-----v-----+
| User DB   |        | Order DB  |        | Inventory |
|           |        |           |        | DB        |
+-----------+        +-----------+        +-----------+

Services communicate asynchronously via events for eventual consistency.

Components

API Gateway

Nginx or Kong

Single entry point routing requests to appropriate microservices

User Service

Spring Boot / Node.js

Manages user profiles and authentication

Order Service

Spring Boot / Node.js

Handles order creation, updates, and status

Inventory Service

Spring Boot / Node.js

Manages product stock levels and availability

Databases

PostgreSQL per service

Each service owns its own database to ensure data encapsulation

Event Bus

Apache Kafka or RabbitMQ

Enables asynchronous communication and event-driven updates between services

Request Flow

1. Client sends request to API Gateway

2. API Gateway routes request to the appropriate microservice

3. Microservice processes request using its own database

4. If needed, microservice publishes events to Event Bus

5. Other services subscribe to relevant events and update their state asynchronously

6. Microservice returns response to API Gateway

7. API Gateway sends response back to client

Database Schema

User Service: User(id PK, name, email, password_hash) Order Service: Order(id PK, user_id FK, status, total_amount) Inventory Service: Product(id PK, name, stock_quantity) Relationships: User Service owns user data; Order Service references user_id but does not modify User data; Inventory Service owns product stock data. No shared databases; communication via events.

Scaling Discussion

Bottlenecks

API Gateway becoming a single point of failure under high load

Database contention within a single service as user base grows

Tight coupling causing cascading failures between services

Event bus overload with high event volume

Data consistency challenges across services

Solutions

Deploy multiple API Gateway instances behind a load balancer

Use database sharding or read replicas per service

Design services with clear boundaries and fallback mechanisms

Partition event topics and scale event bus clusters horizontally

Implement eventual consistency with compensating transactions and sagas

Interview Tips

Time: Spend 10 minutes understanding requirements and clarifying scope, 20 minutes designing service boundaries and data flow, 10 minutes discussing scaling and trade-offs, 5 minutes summarizing.

Explain how you identify bounded contexts from business domains

Discuss importance of data ownership per service

Highlight asynchronous communication benefits and challenges

Mention patterns like Saga for distributed transactions

Address scaling bottlenecks and mitigation strategies

Practice

1. Which of the following best describes the primary goal when identifying service boundaries in microservices?

easy

A. Create services based on the number of developers available

B. Split services evenly by code size

C. Group all database operations into one service

D. Divide the system based on business capabilities and data ownership

Solution

1. Step 1: Understand the purpose of service boundaries

   Service boundaries should reflect business capabilities to ensure clear ownership and independent deployment.

2. Step 2: Evaluate the options

   Options B, C, and D focus on technical or team size factors, which are less effective than business-driven boundaries.

3. Final Answer:

   Divide the system based on business capabilities and data ownership -> Option D

4. Quick Check:

   Business capabilities = A [OK]

Hint: Match services to business functions, not code size or teams [OK]

Common Mistakes:

• Splitting services by code size only
• Grouping all database logic in one service
• Ignoring business domain boundaries

2. Which of the following is the correct way to define a microservice boundary?

easy

A. A service that handles user authentication and profile management

B. A service that mixes payment processing and product catalog updates

C. A service that only manages database connections

D. A service that handles logging for all other services

Solution

1. Step 1: Identify cohesive responsibilities

   A good service boundary groups related business functions, like authentication and profile management.

2. Step 2: Check for unrelated responsibilities

   Options A, B, and C mix unrelated concerns or are cross-cutting, which should be separate services or infrastructure.

3. Final Answer:

   A service that handles user authentication and profile management -> Option A

4. Quick Check:

   Cohesive business functions = D [OK]

Hint: Group related business tasks in one service [OK]

Common Mistakes:

• Combining unrelated business functions
• Creating services for technical concerns only
• Mixing cross-cutting concerns inside business services

3. Given a system with services: OrderService managing orders, InventoryService managing stock, and PaymentService handling payments, which service boundary violation is shown if OrderService directly updates stock quantities?

medium

A. OrderService is violating the single responsibility principle by managing inventory data

B. OrderService is correctly handling all order-related data including stock

C. PaymentService should update stock quantities instead

D. InventoryService should not exist separately from OrderService

Solution

1. Step 1: Analyze service responsibilities

   OrderService should focus on orders; InventoryService owns stock data and updates.

2. Step 2: Identify boundary violation

   OrderService updating stock breaks clear ownership and single responsibility principles.

3. Final Answer:

   OrderService is violating the single responsibility principle by managing inventory data -> Option A

4. Quick Check:

   Single responsibility violation = B [OK]

Hint: Each service owns its data; no direct updates outside boundaries [OK]

Common Mistakes:

• Allowing services to update data owned by others
• Confusing payment service role
• Merging unrelated services unnecessarily

4. A team notices that their UserService and NotificationService are tightly coupled because UserService calls NotificationService directly for every user update. What is the best way to fix this boundary issue?

medium

A. Make NotificationService call UserService instead

B. Merge both services into one to avoid communication

C. Use an event-driven approach where UserService emits events and NotificationService listens

D. Remove NotificationService and handle notifications inside UserService

Solution

1. Step 1: Understand tight coupling problem

   Direct calls create dependencies that reduce service independence.

2. Step 2: Apply event-driven design

   Emitting events decouples services, allowing independent scaling and deployment.

3. Final Answer:

   Use an event-driven approach where UserService emits events and NotificationService listens -> Option C

4. Quick Check:

   Event-driven decoupling = C [OK]

Hint: Use events to decouple services, not direct calls [OK]

Common Mistakes:

• Merging services unnecessarily
• Reversing call direction without decoupling
• Ignoring decoupling benefits

5. You are designing a microservices system for an e-commerce platform. Which approach best defines service boundaries to maximize team autonomy and scalability?

hard

A. Create services based on technical layers like UI, Business Logic, and Database Access

B. Define services around distinct business domains like Catalog, Orders, Payments, and Shipping, each owning its data and APIs

C. Split services by database tables regardless of business function

D. Group all user-related features into one large service to reduce communication

Solution

1. Step 1: Identify business domain boundaries

   Services aligned with business domains allow clear ownership and independent scaling.

2. Step 2: Avoid technical or data-layer splits

   Splitting by technical layers or tables causes tight coupling and reduces autonomy.

3. Step 3: Consider team autonomy and scalability

   Domain-based services enable teams to work independently and scale services as needed.

4. Final Answer:

   Define services around distinct business domains like Catalog, Orders, Payments, and Shipping, each owning its data and APIs -> Option B

5. Quick Check:

   Domain-driven design = A [OK]

Hint: Align services with business domains for autonomy and scale [OK]

Common Mistakes:

• Splitting by technical layers instead of business domains
• Grouping unrelated features together
• Ignoring data ownership in service design