Service Registry and Service Discovery in Microservice

 Service Registry and Service Discovery are complementary mechanisms that enable microservices to locate and communicate with each other dynamically, without hard-coded network addresses.

---

1. Service Registry

A Service Registry is simply a database of service instances (their network locations and metadata):

• What it stores

  • Service name (e.g. “order-service”)

  • Instance ID (to differentiate multiple copies)

  • Hostname/IP and port

  • Optional metadata (version, region, health status)

• How instances register

  • Self-registration: Each service, on startup, calls the registry’s API to add its own address.

  • Health checks: Registry periodically pings instances (HTTP/heartbeat) and removes unhealthy ones.

• Examples

  • Eureka (Netflix OSS)

  • Consul (HashiCorp)

  • ZooKeeper (Apache)

---

2. Service Discovery

Service Discovery is the process by which a client or another service finds the network address of a target service:

2.1 Client-Side Discovery

1. Client queries the registry for instances of “order-service.”

2. It receives a list of healthy instances (e.g. ["10.0.1.12:8080", "10.0.1.13:8080"]).

3. Client performs client-side load balancing (round-robin, random) among them.

• Pros: Full control of load-balancing logic in the client

• Cons: Every client library needs registry integration

2.2 Server-Side Discovery (via API Gateway)

1. Client calls a single API Gateway endpoint (e.g. api.myapp.com/orders).

2. Gateway queries the registry, picks a healthy instance, and forwards the request.

3. Response is proxied back to the client.

• Pros: Simplifies clients, centralizes routing and security

• Cons: Gateway can become a throttling bottleneck if not autoscaled

---

3. Typical Workflow

1. Startup

   • Service instance boots up and sends a “register” request to the registry.

2. Heartbeat

   • Registry pings each instance at regular intervals to confirm health.

3. Discovery Lookup

   • When Service A needs Service B, it queries the registry’s REST API or via a client-side library.

4. Load Balancing

   • The client or gateway selects one instance and makes the call.

5. Deregistration

   • On shutdown or failure, the instance is removed (gracefully or by timeout).

---

4. Why It Matters

• Scalability: You can add or remove instances without reconfiguring clients.

• Fault-Tolerance: Traffic is automatically routed around unhealthy instances.

• Flexibility: Services can move between hosts or ports without breaking communication.

---

Mastering service registry and discovery is crucial for reliable, scalable microservices architectures—enabling services to find each other dynamically and stay resilient as your system grows.