Hybrid cloud and multi-cloud architectural patterns
This article is the second in a multi-part series that covers hybrid cloud and multicloud deployments, architectural patterns, and network topologies. This article explores common hybrid cloud and multicloud architectural patterns. This article also describes scenarios where these patterns are best suited and provides best practices for implementing them using Google Cloud.

The series contains the following parts:

hybrid multi cloud architecture and practices
Hybrid and Multicloud Architecture Patterns (this article)
Hybrid cloud and multicloud network topologies
Every enterprise has a unique mix of application workloads that place requirements and constraints on building a hybrid or multi-cloud setup. Although you must design and tailor your architecture to meet these constraints and requirements, there are some common patterns you can rely on.

Patterns fall into two categories:

Depends on the distributed deployment mode of the application. These patterns are designed to run an application in the computing environment that is most suitable for the application, taking advantage of different properties and characteristics of the computing environment.
A mode based on application redundancy deployment. In these patterns, you can deploy the same application across multiple computing environments for increased capacity or resiliency.
Distributed deployment mode

When migrating from a classic computing environment to a hybrid or multi-cloud setup, be aware of the limitations imposed by existing applications. You also need to take advantage of the unique capabilities each computing environment offers. These distributed patterns aim to achieve a reasonable balance between two goals.

layered mix

Most applications can be classified as front-end applications or back-end applications.

Front-end applications are exposed directly to end users or devices. As a result, these applications are often performance demanding, and new versions may be released frequently as new features and improvements are developed. Because front-end applications typically rely on back-end applications to store and manage data, front-end applications are often stateless or only manage small amounts of data.
Backend applications typically focus on managing data. The main challenges for such applications include processing data in batches and securing it properly. New versions of back-end applications tend to be rolled out less frequently than front-end applications.
The idea behind the layered hybrid model is to focus first on deploying existing front-end applications to the public cloud. In this pattern, you can reuse existing backend applications that remain in their private computing environment. You can migrate front-end applications on a case-by-case basis.