Compute Flashcards
1
Q
What is an Azure Virtual Machine (VM)?
A
- VMs provide IaaS in the form of a virtualised service
- Customer can customise all of the software running on the VM (OS, custom software, custom hosting configs)
- Customer needs to configure, update and maintain all software
- Azure provides preconfigured VM images to allow for rapid provisioning
2
Q
Scaling VMs
A
- VM Scale Sets: allow for the creation and management of identical, load-balanced VMs. Load balancer automatically deployed with VM Set so that resources are being used efficiently
- VM Availability Sets: ensures that VMs stagger updates and have varied power and network connectivity, preventing the loss of all VMs with a single network or power failure.
3
Q
Examples of when to use VMs
A
- Testing and development: quickly create and tear down different OS and application configurations.
- Run applications in the cloud: as opposed to creating a traditional infrastructure. Allows for handling fluctuations in demand.
- Extending a datacenter to the cloud
- During DR: If a primary datacenter fails, a customer can create VMs to run critical applications then shut down the VMs when the datacenter becomes available again
- Lift-and-shift migrations to the cloud
4
Q
VM resources
A
When provisioning a VM, the customer picks the following resources:
- Size (number of processor cores, amount of RAM etc)
- Storage disks (hard disk drives, solid state drives etc)
- Networking (virtual network, public IP address, port configurations)
5
Q
Azure virtual desktop
A
- Another type of VM: a desktop and application virtualisation service that runs on cloud
- Provides centralised security for users’ desktops with Microsoft Entra ID.
- Data and apps are separated from the local hardware. The actual desktop and apps are running in the cloud, meaning the risk of confidential data being left on a personal device is reduced.
6
Q
What are containers?
A
- A virtualisation environment that can be run on a single physical or virtual host.
- Unlike virtual machines, you don’t manage the OS for a container
- Containers are lightweight (much lighter than VMs) and designed to be created, scaled out and stopped dynamically
7
Q
Azure Container Instances
A
- Simplest way to run a container in Azure
- Don’t have to manage VMs
- PaaS offering
- Upload containers and the service runs them
8
Q
Azure Container Apps
A
Similar to Azure Container Instances but allow the ability to incorporate load balancing and scaling (more elasticity)
9
Q
Azure Kubernetes Service
A
AKS is a container orchestration service.
- Manages the lifecycle of containers
10
Q
When to use containers
A
- To achieve a microservice architecture (solutions are broken into smaller, independent pieces)
- E.g. split a website into a container for the front end, another for the backend, another for storage. If the back-end reaches capacity, you could scale it separately to the rest of the application.
11
Q
Azure Functions
A
- Event-driven, serverless compute that doesn’t require maintaining VMs or containers.
- Does not require anything to be “running”: an event wakes the function and runs it
- Use when only concerned about running the application code and not the underlying infra
- Response to an event (e.g. a REST request)
- Scale automatically based on demand
- Runs code when triggered, then deallocates resources when finished, so customer is only charged for CPU when the function runs
- Stateless functions: behave as if they restart every time
- Stateful functions: context is passed through the function to track prior activity
12
Q
Azure App Service
A
- Allows customer to build and host certain types of applications in any programming language without managing infra
- App Service handles deployment & management, securing endpoints, scaling, load-balancing and high availability.
- Four types of App Service:
- Web Apps
- API Apps
- WebJobs
- Mobile Apps
