Book Image

Python Architecture Patterns

By : Jaime Buelta
Book Image

Python Architecture Patterns

By: Jaime Buelta

Overview of this book

Developing large-scale systems that continuously grow in scale and complexity requires a thorough understanding of how software projects should be implemented. Software developers, architects, and technical management teams rely on high-level software design patterns such as microservices architecture, event-driven architecture, and the strategic patterns prescribed by domain-driven design (DDD) to make their work easier. This book covers these proven architecture design patterns with a forward-looking approach to help Python developers manage application complexity—and get the most value out of their test suites. Starting with the initial stages of design, you will learn about the main blocks and mental flow to use at the start of a project. The book covers various architectural patterns like microservices, web services, and event-driven structures and how to choose the one best suited to your project. Establishing a foundation of required concepts, you will progress into development, debugging, and testing to produce high-quality code that is ready for deployment. You will learn about ongoing operations on how to continue the task after the system is deployed to end users, as the software development lifecycle is never finished. By the end of this Python book, you will have developed "architectural thinking": a different way of approaching software design, including making changes to ongoing systems.
Table of Contents (23 chapters)
2
Part I: Design
6
Part II: Architectural Patterns
12
Part III: Implementation
15
Part IV: Ongoing operations
21
Other Books You May Enjoy
22
Index

Scheduled downtime

While ideally there should be no interruption in the system as a result of the changes made, sometimes it's simply not possible to perform big changes without interrupting the system.

When and whether it's sensible to have downtime may depend greatly depending on the system. For example, in its first years of operation, the popular website Stack Overflow (https://stackoverflow.com/) had frequent downtime, initially even every day, where the webpage returned a "down for maintenance" page during the morning hours in Europe. That changed eventually, and now it's rare to see that kind of message.

But that was acceptable in the early stages of the project as the bulk of their users used the site in line with North American hours and it was (and still is) a free website.

Scheduling downtime is always an option, but it's a costly one, so it needs to be designed in a way that minimizes the impact on the operations...