Microsoft Azure Well-Architected Framework - Performance efficiency

https://docs.microsoft.com/en-us/learn/modules/azure-well-architected-performance-efficiency/2-scaling-up-and-scaling-out

What is scaling? When we look at ways to increase or decrease the compute capacity and relative costs for your applications, it's important to define two key concepts: scaling and resources.

Scaling is the process of managing your resources to help your application meet a set of performance requirements. When you have too many resources serving your users, you won't be using those resources efficiently, and you'll be wasting money. When you have too few resources available, the performance of your application can be adversely affected. Your goal is to meet your defined performance requirements while optimizing for cost. Resources can refer to anything you need to manage and run your applications. Memory and CPUs for virtual machines are the most obvious resources. But some Azure services might require you to consider bandwidth or abstractions, like Request Units (RUs) for Azure Cosmos DB.

Easy scaling is a key benefit of Azure. Most Azure resources let you easily add or remove resources as demand changes. Many services have automated options that monitor demand and adjust for you. This automatic scaling capability is commonly known as autoscaling.

Let's look at what scaling up or down means in the context of Azure resources:

In Azure virtual machines, you scale based on a virtual machine size. Each VM size has a certain amount of vCPUs, RAM, and local storage associated with it. For example, you could scale up from a Standard_DS1_v2 virtual machine (1 vCPU and 3.5 GB of RAM) to a Standard_DS2_v2 virtual machine (2 vCPUs and 7 GB of RAM).

Azure SQL Database is a platform as a service (PaaS) implementation of Microsoft SQL Server. You can scale up a database based on the number of database transaction units (DTUs) or vCPUs. DTUs are an abstraction of underlying resources and are a blend of CPU, IO, and memory. For example, you could scale your database in Azure SQL Database from a size of P2 with 250 DTUs up to a P4 with 500 DTUs to give the database more throughput and capacity.

Azure App Service is a PaaS website-hosting service on Azure. Websites run on a virtual server farm, which is also known as an App Service plan. You can scale the App Service plan up or down between tiers. You also have capacity options within those tiers. For example, an S1 App Service plan has 1 vCPU and 1.75 GB of RAM per instance. You could scale up to an S2 App Service plan, which has 2 vCPUs and 3 GB of RAM per instance.

Scaling out is the process of adding more instances to support the load of your solution

Scaling in is the process of removing instances that are no longer needed to support the load of your solution

# Autoscale

You can configure some of these services to use a feature called autoscale. With autoscale, you no longer have to worry about scaling services manually. Instead, you can set a minimum and maximum threshold of instances.

# Throttling

You can also use a throttling mechanism to limit the number of requests from a source. You can safeguard performance limits by putting known limits into place at the application level.

Serverless

Serverless computing provides a cloud-hosted execution environment that runs your apps but completely abstracts the underlying environment. When you work with serverless apps, infrastructure isn't your responsibility. Scaling and performance are handled automatically. You're only billed for the resources you use.

Containers

Containers are lightweight and well suited to scale-out scenarios. They're designed to be created, scaled out, and stopped dynamically as your environment and demands change. Another benefit of using containers is the ability to run multiple isolated applications on each virtual machine. Because containers are secured and isolated at a kernel level, you don't necessarily need to separate VMs for separate workloads.

Although you can run containers on virtual machines, there are a couple of Azure services that ease the management and scaling of containers:

Azure Kubernetes Service (AKS)

With Azure Kubernetes Service, you can set up virtual machines to act as your nodes. Azure hosts the Kubernetes management plane. You're charged only for the running worker nodes that host your containers.

Azure Container Instances

Azure Container Instances is a serverless approach that lets you create and execute containers on demand. You're charged only for the execution time per second.