You may think that your application is already cloud-ready, it's deployed in the cloud isn't it?!
But in truth, if your application is hard to monitor/scale/release/maintain, lacks resiliency, and is poorly structured, then it's not "cloud-ready". So what does "cloud-ready" actually mean?
A “cloud ready” application is a legacy software application that has been modified to run on cloud computing infrastructure. Whether you are looking at a modular monolith or a full-blown distributed system, building a cloud-ready, resilient application is a must.
In this workshop, we will look at the requirements for Tacky Tacos. A fictional taco business that's thriving and wishes to expand. We'll look at all the different requirements that need to be met, starting with a code kata and how we can apply various design patterns to our way of thinking. We'll then move on to actually building out a functioning, all-be-it, contrived application in .NET 7 and C#, learning as we go.
Topics we'll cover:
Architecture and Methodology
-
Modular monolith vs microservices vs SOA
-
The "12 Factors" and what they are
-
Distribution and Modularisation
-
Common design patterns
-
Modelling requirements
-
Practical: Code kata for tacky tacos
Implementations
-
.NET 7 application development recap
-
Dependency Injection
-
HttpClientFactory
-
Good practices
-
External Configuration
-
What it is and why it is useful
-
Practical: implement external configuration with
-
Databases
-
Why separate DBs may be required
-
Eventual Consistency
-
DB options - SQL, Document, Table, Blob, Serverless
-
Practical: Create two distinct databases - later these will employ "eventual consistency"
-
Synchronous communication
-
Introduction to synchronous communication and when to use it
-
Synchronous options
-
Practical: build an HTTP endpoint for a web client
-
Practical: implement gRPC to communicate with a non .NET application
-
Practical: implement a hub with SignalR to broadcast real-time information
-
Asynchronous communication
-
Introduction to asynchronous communication and when to use it
-
Message brokers
-
Implementation considerations
-
Error handling
-
Choosing a message broker
-
Practical: Implement a pub/sub with RabbitMQ
-
API gateways and BFF
-
Introduction to the use cases and considerations
-
When to use BFF
-
Practical: API Gateway with YARP
-
Load balancing
-
What it is and why we need it
-
Practical: Add load balancing to YARP
-
Service discovery
-
What it is and why we need it
-
Practical: set up Service Discovery using Steeltoe and Eureka
-
Practical: Integrate Service Discovery with YARP API gateway
-
Resiliency
-
What makes an application resilient?
-
policies, circuit breakers, fallbacks and retries
-
Practical: implement policies with Polly
-
M2M Authentication
-
Why it's needed
-
Practical: Implement M2M authorization with Auth0
-
Monitoring, tracing, observability
-
What is it and why do we need it
-
What is the OpenTelemetry project
-
What tools are available
-
Practical: Implement various endpoints and actuators with Steeltoe
-
Serverless
-
When to use a serverless function and why
-
Practical: create an Azure Function service
