Structured review of Azure architectures: A guide for web application review

 Invest a lot of effort to create a perfect architecture before construction should be avoided because this is unrealistic in software development. It is recommend investing in a little bit of architecture envisioning early to identify architecture options in high level, but details should emerge over time. Development team should ensure that architecture will be validated through functioning software, and will be reviewed frequently throughout the project to support optimal implementation. Moreover, emergent architecture should influence and correct intentional architecture.

Architecture review should be performed through out the project development. If you have web app work load in Azure or if you are thinking to move it to Azure, its recommend to perform structured architecture review on following key areas:

• Availability
• Scalability
• Resiliency
• DevOps
• Security
• Management Tools

Performing an Azure architecture review is no longer an activity limited to experts. This paper, by Mahesh Kshirsagar of the AzureCAT team, identifies review drivers and explains how to evaluate your current architecture against these drivers to detect and address current risks. Project teams of any size can self-start a review with this resource. Click download button below to get free PDF copy.

Click here to Download Guide

Ref: Microsoft Docs.

Choosing Compute Services on Azure (AKS, Service Fabric, App Service)

If you are planning to move your services to cloud its important to understand what is the best candidate compute service for you. You don't want to over/under  provision your solution. So what is the best service we should pick, this is on of the question that i am asked often. While there is no single right or wrong answer it depends on what kind of workload and SLAs you have that will dictate what is best compute service for you.

Azure Compute Services offer following key workloads:

• App Service: A managed service for hosting web apps, mobile app back ends, RESTful APIs, or automated business processes.

• Azure Kubernetes Service (AKS):  A managed Kubernetes service for running containerized applications.

• Batch : A managed service for running large-scale parallel and high-performance computing (HPC) applications 

• Container Instances: The fastest and simplest way to run a container in Azure, without having to provision any virtual machines and without having to adopt a higher-level service. 

• Functions: A managed FaaS service. 

• Service Fabric: A distributed systems platform that can run in many environments, including Azure or on premises. 

• Virtual machines: Deploy and manage VMs inside an Azure virtual network.

It is also important to understand following hosting models:

• Infrastructure-as-a-Service (IaaS) lets you provision individual VMs along with the associated networking and storage components. Then you deploy whatever software and applications you want onto those VMs. This model is the closest to a traditional on-premises environment, except that Microsoft manages the infrastructure. You still manage the individual VMs. 

• Platform-as-a-Service (PaaS) provides a managed hosting environment, where you can deploy your application without needing to manage VMs or networking resources. Azure App Service is a PaaS service. 

• Functions-as-a-Service (FaaS) goes even further in removing the need to worry about the hosting environment. In a FaaS model, you simply deploy your code and the service automatically runs it. Azure Functions is a FaaS service

 Following chart show how best to pick candidate compute service for you:

You also need to look at aspects such as scalability, availability, security and how easily you can perform DevOps.

Ref: Microsoft Docs.

What is Bicep

Have you been writing ARM templates and didn't quite fancy the way ARM templates are written in JSON. Well Bicep is here that would help you author ARM templates with much cleaner syntax. Bicep give you abstraction the way ARM templates are now written. Its supported by Microsoft 100% free, very modular, state is stored in Azure no manual handling. 

You can get latest windows installer here or see all install options here: bicep/installing.md at main · Azure/bicep (github.com)

You can also download Bicep Visual Studio Code extension.

Hello world sample Bicep file to spin up storage in Azure..

param location string = 'eastus'

@minLength(3)
@maxLength(24)
param storageAccountName string = 'azmubistorageacc1' // must be globally unique

var storageSku = 'Standard_LRS' // declare variable and assign value

resource stg 'Microsoft.Storage/storageAccounts@2019-06-01' = {
  name: storageAccountName
  location: location
  kind: 'Storage'
  sku: {
    name: storageSku // reference variable
  }
}

output storageId string = stg.id // output resourceId of storage account

Edit it in VS Code using Bicep extensions.

Deploy template using Bicep CLI (make sure to create resource group in Azure in advance e.g. bicep).

az deployment group create -f ./test.bicep -g bicep

Storage account created in Azure using Bicep.

Bicep Language specifications can be found here.

Bicep Azure DevOps Task is available here.

steps:
- task: BicepBuild@0
  inputs:
    process: 'single'
    sourceFile: '.\bicep_files\sample1.bicep'
    stdout: false # Note if stdout is true 'outputDirectory' will not be interpreted
    outputFile: '.\bicep_files\sample1.out.json' # Only when 'stdout' is false or not defined and 'outputDirectory' is empty or not defined

Bicep Build Actions (Github Action) is available to run the Bicep CLI to build ARM template more.

steps:
# Runs the bicep CLI action - individual files
- name: Run Bicep build
  uses: aliencube/bicep-build-actions@v0.3
  with:
    files: sample1.bicep sample2.bicep biceps/sample3.bicep biceps/sample4.bicep

# Checks the result
- name: Check the result
  shell: bash
  run: |
    shopt -s globstar
    ls -altR **/sample*.*

If you have existing ARM templates you can decompile them.

bicep decompile "path/to/MyARMTempateFile.json"

You can also export your Azure resource group to .bicep file.

step 1 - az group export --name "MY-Azure-Resource-Group" > MainARMTemplate.json
step 2 - bicep decompile MainARMTemplate.json

Learn more about Bicep here ..

Azure Traffic Manager vs App Gateway vs Front Door vs Load Balancer

 

Some of the key features and difference between Azure Traffic Manager, App Gateway, Front Door and Load Balancer.

Traffic Routing methods.

Traffic Manager:

• Priority (default and backup endpoints)
• Weighted (distribute across endpoints)
• Performance (nearest endpoint)
• Geographic (geography controlled for GDPR etc)
• Subnet (endpoints base don IP)
• Multi-value (multiple ends available)

Front Door:

• Latency (least latency endpoint)
• Priority (primary then secondary)
• Weighted (based on endpoint weight)
• Session Affinity (same endpoint per session)

Application Gateway:

• Multiple site hosting
• URL routing
• Redirection
• Rewrite HTTP headers and URL

Decision Tree:

  

Reference Architecture Examples:

The following table lists various architecture reference articles based on the load-balancing services used as a solution.

Service(s)	Article	Description
Load Balancer	Load balance virtual machines (VMs) across availability zones	Load balance VMs across availability zones helps to protect your apps and data from an unlikely failure or loss of an entire datacenter. With zone-redundancy, one or more availability zones can fail and the data path survives as long as one zone in the region remains healthy.
Front Door	Sharing location in real time using low-cost serverless Azure services	Use Azure Front Door to provide higher availability for your applications than deploying to a single region. If a regional outage affects the primary region, you can use Front Door to fail over to the secondary region.
Application Gateway	IaaS: Web application with relational database	Learn how to use resources spread across multiple zones to provide a high availability (HA) architecture for hosting an Infrastructure as a Service (IaaS) web application and SQL Server database.
Traffic Manager	Multi-tier web application built for high availability and disaster recovery	Deploy resilient multi-tier applications built for high availability and disaster recovery. If the primary region becomes unavailable, Traffic Manager fails over to the secondary region.
Azure Front Door + Application Gateway	Multitenant SaaS on Azure	Use a multi-tenant solution that includes a combination of Front Door and Application Gateway. Front Door helps load balance traffic across regions and Application Gateway routes and load-balances traffic internally in the application to the various services that satisfy client business needs.
Traffic Manager + Load Balancer	Multi-region N-tier application	A multi-region N-tier application that uses Traffic Manager to route incoming requests to a primary region and if that region becomes unavailable, Traffic Manager fails over to the secondary region.
Traffic Manager + Application Gateway	Multi-region load balancing with Traffic Manager and Application Gateway	Learn how to serve web workloads and deploy resilient multi-tier applications in multiple Azure regions, in order to achieve high availability and a robust disaster recovery infrastructure.