4+1 Architectural View Model

It’s incredibly difficult to describe a complex system, regardless if you are developing a monolith or (micro) services. Use cases, code organization/navigation, interactions between services, and deployment/infrastructure are just some of the aspects that comprise the architecture of an entire system.

Depending on your role as a stakeholder, your view of the system can be very different than another stakeholder.

This blog post is in a series. To catch up check out these other posts:

Context Matters

There are many different stakeholders related to a software system, which all have different perspectives. Project/Project Managers, Developers, System Engineers, End Users all view a system in completely different ways. They view the system based on their own context.

In order to describe a system, it would be useful to define all the different viewpoints and how the overall use cases of the system.

4+1 Architectural View Model

We all have seen many books and articles where one diagram attempts to capture the gist of the architecture of a system. But looking carefully at the set of boxes and arrows shown on these diagrams, it becomes clear that their authors have struggled hard to represent more on one blueprint than it can actually express. Are the boxes representing running programs? Or chunks of source code? Or physical computers? Or merely logical groupings of functionality? Are the arrows representing compilation dependencies? Or control flows? Or data flows? Usually it is a bit of everything.

The paper by Philippe Kruchten, Architectural Blueprints—The “4+1” View Model of Software Architecture, defines 4 concurrent views from the point of view of the various stakeholders.

4+1 Architecture View Model

I recommend reading the paper but for an incredibly simplified version of the views:

  • Logical View: The functionality. The service.
  • Process View: Communication between processes and/or services.
  • Physical View: Deployment of your services.
  • Development View: File/Folder Structure of your codebase. What you’re looking at in your IDE/Editor

The +1 comes in from the scenarios view which is what your end users actually care about. It’s the system functionality/capabilities. The scenarios view is what guides all the other views.

Service Boundaries

Where I think general guidance has fallen short when developing (micro)services is just because you’ve identified multiple logical views ( services), does not mean that each must have their own independent deployment (physical view) or git repo (development view).

To be clear, there are obvious benefits to having independent deployments, but there are many disadvantages as well. But it’s not a requirement.

As an example, in .NET, you could have a solution with various projects that represent different logical views. None of these projects reference each other, they are simply in the same solution. There may however be an ASP.NET Core project that hosts these projects as a single process.

Point being is that just because you have multiple services does not necessarily mean they need to be developed independently or deployed independently.

The four views are representations of the system depending on the context. They don’t need to map 1:1.

When not everything maps 1:1, that’s when it makes it a bit more challenging finding service boundaries. Focus on the scenarios.

Blog Series

More on all of these topics will be covered in greater length in other posts. If you have any questions or comments, please reach out to me in the comments section or on Twitter.

Enjoy this post? Subscribe!

Subscribe to our weekly Newsletter and stay tuned.

Roundup #48: .NET Reunified, Performance in .NET Core 3, XAML Hot Reload, Blazor

Here are the things that caught my eye recently in .NET.  I’d love to hear what you found most interesting this week.  Let me know in the comments or on Twitter.

.NET Reunified: Microsoft’s Plans for .NET 5

When Microsoft announced .NET 5 at Microsoft Build 2019 in May, it marked an important step forward for developers working across desktop, Web, mobile, cloud and device platforms. In fact, .NET 5 is that rare platform update that unifies divergent frameworks, reduces code complexity and significantly advances cross-platform reach.

Link: https://msdn.microsoft.com/en-us/magazine/mt833477.aspx

What’s new for performance in .NET Core and ASP.NET Core 3.0 – Ben Adams

One of the biggest advantages of using .NET Core (besides cross-platform support) is the drastic improvements in performance. Because the .NET Core team was able to make minor breaking changes in the runtime and Base Class Library (BCL), lots of stuff was implemented much more efficiently. In this session Ben will dive into the performance improvements in .NET Core in the 3.0 release: runtime changes, JIT changes, intrinsics and a deep dive into some of the improvements making it the best release yet!

Link: https://www.youtube.com/watch?v=bIfolJJJWqs

Announcing XAML Hot Reload for Xamarin.Forms

Today at Xamarin Developer Summit, we announced XAML Hot Reload for Xamarin.Forms, which enables you to make changes to your XAML UI and see them reflected live, without requiring another build and deploy.

XAML Hot Reload for Xamarin.Forms speeds up your development and makes it easier to build, experiment, and iterate on your user interface. And this means that you no longer have to rebuild your app each time you tweak your UI – it instantly shows you your changes in your running app!

Link: https://devblogs.microsoft.com/xamarin/xaml-hot-reload/

Blazor, a new framework for browser-based .NET apps – Steve Sanderson

Today, nearly all browser-based apps are written in JavaScript (or similar languages that transpile to it). That’s fine, but there’s no good reason to limit our industry to basically one language when so many powerful and mature alternate languages and programming platforms exist. Starting now, WebAssembly opens the floodgates to new choices, and one of the first realistic options may be .NET.

Blazor is a new experimental web UI framework from the ASP.NET team that aims to brings .NET applications into all browsers (including mobile) via WebAssembly. It allows you to build true full-stack .NET applications, sharing code across server and client, with no need for transpilation or plugins.

In this talk I’ll demonstrate what you can do with Blazor today and how it works on the underlying WebAssembly runtime behind the scenes. You’ll see its modern, component-based architecture (inspired by modern SPA frameworks) at work as we use it to build a responsive client-side UI. I’ll cover both basic and advanced scenarios using Blazor’s components, router, DI system, JavaScript interop, and more.

Link: https://www.youtube.com/watch?v=uW-Kk7Qpv5U

Enjoy this post? Subscribe!

Subscribe to our weekly Newsletter and stay tuned.

Roundup #47: Elastic APM, Miguel de Icaza, Microsoft.FeatureManagement, .NET Core 3 Progress

Here are the things that caught my eye recently in .NET.  I’d love to hear what you found most interesting this week.  Let me know in the comments or on Twitter.

Elastic APM .NET Agent beta Released

We are proud to announce the beta release of the Elastic APM .NET agent! The version number of this release is 1.0.0-beta1.

Back in February we announced the alpha release of the .NET APM agent. We received very positive feedback from the community, and we managed to attract lots of users already in this very early stage. Our base NuGet package reached around 20,000 downloads.

Link: https://www.elastic.co/blog/elastic-apm-dot-net-agent-beta-released

Careers Behind the Code: Building Careers and Companies with Open Source with Miguel de Icaza

In this episode, we talk to Miguel de Icaza, Distinguished Engineer about his career building open source software, founding companies, and building communities.

Link: https://www.youtube.com/watch?v=ApLBy0AJBEw

Microsoft.FeatureManagement

In a recent .NET Community Standup, a new library was introduced that’s being built by the Azure team – Microsoft.FeatureManagement. In this post, I give a brief introduction to the library and how to use it in an ASP.NET Core app. This post just covers the basics – in later posts I’ll show some of the ASP.NET Core-specific features, as well as how to create custom feature filters.

Link: https://andrewlock.net/introducing-the-microsoft-featuremanagement-library-adding-feature-flags-to-an-asp-net-core-app-part-1/

.NET Core 3.0 Progress

Link: https://twitter.com/ziki_cz/status/1146553003610202112

Enjoy this post? Subscribe!

Subscribe to our weekly Newsletter and stay tuned.