Microservices Architecture: Principles, Characteristics, and Its Relationship with DevOps

Microservices

Microservice Architecture describes a design approach where an application is built as a set of independent, deployable services. Although the term lacks a precise definition, common traits include business‑capability‑oriented organization, automated deployment, endpoint intelligence, and language‑ and data‑agnostic boundaries.

Microservice Style Characteristics

Microservices share several informal characteristics: each service runs in its own process, communicates via lightweight mechanisms (typically HTTP APIs), is built around a specific business capability, and can be deployed independently. Services may be written in different languages and use different data stores, minimizing centralized management.

Componentization and Services

A component is a replaceable, upgradable software unit. In microservice architectures, components become out‑of‑process services that communicate via RPC or web service calls, allowing independent deployment and reducing the need to rebuild an entire monolith for a single change.

Organization Around Business Capabilities

Teams are organized cross‑functionally around business functions rather than technical layers, reflecting Conway's Law: the system structure mirrors the communication structure of the organization. This alignment encourages clear service boundaries and reduces coordination overhead.

Product vs. Project

Microservice advocates argue that teams should own the product for its entire lifecycle ("you build, you run it"), fostering continuous responsibility, faster feedback, and stronger alignment with business goals.

Strengthening Endpoints and Weakening the Bus

Microservices favor simple, REST‑style HTTP APIs or lightweight messaging over heavyweight ESB solutions, promoting loose coupling and high cohesion.

Decentralized Governance

Instead of a single platform standard, teams choose tools that fit their problems (e.g., Node.js for reporting, C++ for high‑performance components). Polyglot persistence and distributed data management become common, with each service managing its own database.

Infrastructure Automation

Cloud platforms (especially AWS) and CI/CD pipelines have reduced the complexity of building, testing, and deploying microservices. Automated testing, continuous deployment, and modular deployment pipelines enable rapid, repeatable releases.

Fault‑Tolerance Design

Microservice systems must tolerate service failures. Practices such as circuit breakers, health checks, and automated fault injection (e.g., Netflix's Simian Army) provide early detection and recovery mechanisms.

Design Evolution

Microservice teams continuously refine service boundaries, favoring replaceable components and avoiding long‑lived monolithic services. Incremental redesigns allow faster iteration while keeping the overall system stable.

Is Microservices the Future?

The article concludes that microservices offer compelling benefits—independent deployment, scalability, and alignment with DevOps—but acknowledges that the approach is still maturing, and its long‑term suitability depends on team skill, organizational culture, and the ability to manage distributed complexity.

Other Topics

How Large Are Microservice Systems?

Service size varies widely; teams often follow the "two‑pizza" rule (no more than ~20 people per service), but smaller teams can also own services.

Microservices vs. SOA

While microservices share some goals with Service‑Oriented Architecture, they differ in implementation focus, avoiding heavyweight ESBs and emphasizing lightweight protocols and decentralized governance.

Multi‑Language, Multi‑Choice

Microservice environments encourage polyglot development, allowing different services to be written in the most suitable language for their domain.

Standards and Open Protocols

Microservice teams prefer open, internet‑scale standards (HTTP, ATOM, etc.) over proprietary, enforced enterprise standards.

Making the Right Things Easier

Automation tools (e.g., Netflix OSS, Dropwizard) support packaging, code management, and monitoring, simplifying continuous delivery.

Circuit Breaker and Existing Code

Circuit breakers, bulkheads, and timeouts—patterns described in "Release It!"—are essential for building resilient microservice communication.

Synchronous Calls Are Harmful

Synchronous inter‑service calls can amplify downtime; many organizations adopt asynchronous APIs to mitigate this risk.

Please follow the WeChat public account devopshub for more DevOps information.