Foundations
Architecture styles, sometimes called architecture patterns, define how components relate to each other and cover various architectural characteristics. These style names are like shortcuts for experienced architects. For example, if an architect talks about a 'layered monolith,' it signals specific structural aspects, what works well (or not), common deployment models, data strategies, and more.
An architecture style describes the structure and typical architectural characteristics, both good and bad. Architects should also understand the fundamental patterns often found within these styles.
Fundamental Patterns
Over time, some basic patterns keep showing up in software architecture. These patterns help us organize code and deployments. For instance, the idea of separating different concerns based on functionality, into layers is an age-old concept in software. It's been around for a long time and still appears in various forms, even in modern architecture.
Unitary Architecture
In the early days of software, the computer and software were one. But as technology evolved, they separated to handle more complex tasks. Mainframes, for instance, began as single systems but later split data into separate systems. Today, unitary architectures are rare, except in specific situations like embedded systems. Most software systems grow in complexity, so we separate them into different parts to keep them running smoothly.
Client-Server
A fundamental style in architecture separates technical functionality between frontend and backend. This configuration is commonly referred to as a two-tier or client/server architecture. Over the years, this architectural approach has taken on various forms, adapting to different eras and advancements in computing technology.
Desktop + Database Server
In the past, desktop applications used to handle the user interface, while the data was stored in separate database servers accessible over networks. This allowed desktops to handle user interfaces while heavy data processing happened on more powerful database servers.
Browser + Web Server
In this setup, users interact with a web server, which connects to a database server. It's similar to the desktop architecture but with lighter clients (browsers), making it more accessible. Although the database is separate, it's still considered two-tier because the web and database servers run on one machine in the data center, while the user interface is in the user's browser.
Three-tier
A popular architecture from the late 1990s is the three-tier architecture, which adds more layers of separation. It includes a database tier with a powerful server, an application tier managed by an application server, and a frontend with HTML and JavaScript for user interfaces.