Book Image

ASP.NET Core 2 High Performance - Second Edition

By : James Singleton
Book Image

ASP.NET Core 2 High Performance - Second Edition

By: James Singleton

Overview of this book

The ASP.NET Core 2 framework is used to develop high-performance and cross-platform web applications. It is built on .NET Core 2 and includes significantly more framework APIs than version 1. This book addresses high-level performance improvement techniques. It starts by showing you how to locate and measure problems and then shows you how to solve some of the most common ones. Next, it shows you how to get started with ASP.NET Core 2 on Windows, Mac, Linux, and with Docker containers. The book illustrates what problems can occur as latency increases when deploying to a cloud infrastructure. It also shows you how to optimize C# code and choose the best data structures for the job. It covers new features in C# 6 and 7, along with parallel programming and distributed architectures. By the end of this book, you will be fixing latency issues and optimizing performance problems, but you will also know how this affects the complexity and maintenance of your application. Finally, we will explore a few highly advanced techniques for further optimization.
Table of Contents (20 chapters)
Title Page
Credits
Foreword
About the Author
About the Reviewer
www.PacktPub.com
Customer Feedback
Preface
3
Setting Up Your Environment
4
Measuring Performance Bottlenecks

Summary


In this chapter, we discussed some techniques that can improve the performance of code execution and dug into the projects that make up .NET Core and ASP.NET Core. We explored data structures, serialization, hashing, parallel programming, and how to benchmark to measure relative performance. We also covered how to perform multithreading, concurrency, and locking with C#.

Linear performance characteristics are easier to scale and code that does not exhibit this behavior can be slow when the load increases. Code that has an exponential performance characteristic or has erratic outliers (which are rare but very slow when they occur) can cause performance headaches. It is often better to aim for code that, while being slightly slower in normal cases, is more predictable and performs consistently over a large range of loads.

The main lesson here is to not blindly apply parallel programming and other potentially performance-enhancing techniques. Always test to make sure that they make a...