Service Discovery Client/Server Side

🔎 What is Service Discovery?

In microservices, services need to find and call each other.

But:

• IP addresses change

• Containers restart

• Pods scale up/down

So instead of hardcoding IPs, we use service discovery.

There are two main types:

• Client-side discovery

• Server-side discovery

🧭 1️⃣ Client-Side Discovery

📌 How It Works

The client asks a service registry:

“Where is Service B?”

The registry returns a list of available instances.

The client:

• Chooses one instance (load balancing)

• Calls it directly

So the client is responsible for:

• Looking up

• Load balancing

• Retry logic

📊 Flow

Client → Service Registry → Gets instance list → Calls Service directly

🛠 Example

Common tools:

• Netflix Eureka

• Consul

• Spring Cloud

In Kubernetes:

• Internal DNS-based service discovery

🏦 Banking Example

Retail Service needs Customer Service.

Steps:

1. Retail queries Eureka

2. Gets 3 instances of Customer Service

3. Retail picks one

4. Direct REST call

✅ When to Use Client-Side Discovery

✔ You control all clients✔ Microservices are built in same tech stack✔ You want fine-grained client control✔ Internal-only architecture

Most internal microservice architectures use this.

🌐 2️⃣ Server-Side Discovery

📌 How It Works

The client does NOT query registry.

Instead:

• Client calls a load balancer

• Load balancer queries registry

• Load balancer forwards request

So discovery happens at infrastructure layer.

📊 Flow

Client → Load Balancer → Registry → Target Service

🛠 Example

• Amazon Web Services ELB

• Microsoft Azure Application Gateway

• NGINX

• Kubernetes Service

🏦 Banking Example

Mobile App → API Gateway → Internal Services

The mobile app doesn’t know anything about service instances.

API Gateway:

• Routes

• Load balances

• Performs discovery

✅ When to Use Server-Side Discovery

✔ External clients (mobile/web)✔ Polyglot clients (multiple tech stacks)✔ You want centralized routing✔ You want better security control✔ Large enterprise environments

Most banks use server-side discovery for external traffic.

⚖️ Key Differences

Feature	Client-Side	Server-Side
Who does discovery?	Client	Load balancer
Load balancing	Client	Infrastructure
Complexity in client	Higher	Lower
Best for	Internal services	External traffic
Control	Distributed	Centralized

🧠 Enterprise Architect Perspective

In a bank:

• Internal microservices → Client-side discovery

• External/mobile traffic → Server-side discovery via API Gateway

This gives:

• Flexibility internally

• Control and security externally

🎤

You can say:

“Client-side discovery delegates service lookup and load balancing to the client, making it suitable for internal microservices. Server-side discovery centralizes this responsibility in a load balancer or gateway, making it ideal for external-facing or polyglot environments where centralized control and security are important.”