Multi-tier Architecture

Owned by Computer Science Department
Last updated: Aug 05, 2023
2 min read

In database systems, the concept of 'tiers' typically refers to the architecture design pattern, where different system functionalities are separated into distinct layers, each located on different servers or clusters of servers. This is often done for reasons of scalability, manageability, and performance.

Here are the most common types of tiered architectures:

  1. Single-Tier Architecture: In this simplest form, all the components of the application, including the user interface, business logic, and data storage, all reside on the same machine. It's typically used for small, simple applications or for prototyping.

  2. Two-Tier Architecture: In a two-tier or client-server architecture, the system is divided into two parts: a server (often containing both the business logic and data storage components), and clients (containing the user interface and user input logic). The clients send requests to the server, which processes the requests and returns the results.

  3. Three-Tier Architecture: In a three-tier architecture, the system is divided into three parts, each on a separate platform:

    • The presentation tier (also called the front-end or client tier) is the user interface.

    • The application tier (or middle tier) contains the business logic, and processes commands, performs calculations, and makes logical decisions.

    • The data tier includes the database and the data management components.

This architecture improves scalability, as each tier can be scaled independently. It also improves maintainability and separation of concerns, as changes in one tier don't affect the others.

  1. N-Tier/Multi-Tier Architecture: This is an extension of the three-tier model that can have any number of distinct tiers. The most common use of this architecture includes breaking the business logic layer into separate tiers, for instance, one for web server (handling HTTP requests), another for application server (running application logic), and another for message queues or other specialized services.

Each of these tiers can exist on separate machines, potentially allowing for better performance, scalability, and fault isolation.

In a database context, typically the database would reside in the data tier. In a distributed database system, the data tier itself might be distributed over multiple nodes, either for reasons of data size (sharding), resilience (replication), or both.

However, these are logical separations, and depending on the specific application and infrastructure, multiple tiers may coexist on the same machine or be spread across multiple machines. The choice of architecture depends on many factors, including the complexity and requirements of the specific application, as well as the resources available.