The latest articles on DEV Community by Bulat Gabdrakhmanov (@bulatgab). https://dev.to/bulatgab https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F234830%2F4de561cd-066d-4bd7-9423-11961974bdb7.jpeg https://dev.to/bulatgab en Bulat Gabdrakhmanov Thu, 30 Jan 2025 07:47:14 +0000 https://dev.to/bulatgab/microservices-101-communication-between-services-4ea https://dev.to/bulatgab/microservices-101-communication-between-services-4ea <p>Microservice architecture consists of multiple independent services that often need to communicate with each other. </p> <p>This communication can be achieved in two primary ways:</p> <ul> <li><strong>Synchronous</strong></li> <li><strong>Asyncronous</strong></li> </ul> <h1> Synchronous communication </h1> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fcmo29her749ekznxfxxo.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fcmo29her749ekznxfxxo.png" alt="Image description" width="800" height="196"></a></p> <p>Synchronous communication follows a client-server model, where one service sends a request to another and waits for a response before proceeding.</p> <h2> Common protocols </h2> <ul> <li>REST APIs – resource-based communication using HTTP methods (GET, POST, PUT, DELETE)</li> <li> <p>RPC - <strong>R</strong>emote <strong>P</strong>rocedure <strong>C</strong>all. Allows invoking methods on remote services as if they were local. Examples:</p> <ul> <li>SOAP - XML-based</li> <li>gRPC - uses binary format for payloads, high performance,</li> </ul> </li> </ul> <h2> Pros &amp; Cons </h2> <p><strong>Pros:</strong></p> <ul> <li>Simple to use and implement</li> <li>Immediate data retrieval</li> </ul> <p><strong>Cons:</strong></p> <ul> <li> <strong>Tightly coupled</strong> - if one service depends on another that is down, the entire operation fails</li> <li> <strong>Scalability issues</strong> - high traffic can overload serivces</li> <li> <strong>Latency</strong> - slow response times can create bottlenecks</li> </ul> <h2> Use cases </h2> <ul> <li>External API calls</li> <li>Need immediate response (e.g. authentication)</li> <li>Data consistency is critical (e.g. order processing)</li> </ul> <h1> Asynchronous communication </h1> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F71j5u0dbogil2b8a0ml4.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2F71j5u0dbogil2b8a0ml4.png" alt="Image description" width="800" height="362"></a></p> <p>Asynchronous communication uses message brokers as an intermediate service for communication between services. Instead of waiting for an immediate response, a service sends a message to a broker, which another service later processes.</p> <h2> Pros &amp; Cons </h2> <p><strong>Pros:</strong></p> <ul> <li> <strong>Loosely coupled</strong> - services operate independently, improving scalability</li> <li> <strong>High availability and resilience</strong> - a service can continue operating even if other services are temporarily down</li> </ul> <p><strong>Cons:</strong></p> <ul> <li> <strong>Increased complexity and maintenance</strong> - requires additional infrastructure and message handling</li> <li> <strong>Messages ordering &amp; duplication</strong> - messages may arrive out of order or be duplicated, requiring deduplication strategies</li> </ul> <h2> Use cases </h2> <ul> <li>Event-driven architecture</li> <li>Background processing - immediate response is not required</li> <li>High-load systems - where scalability is a priority</li> </ul> microservices microservices101 backend