Scalability with Load Balancer in Azure

Background

Your organization already has a VM that host web server to the end-user. You want to make the VM can work scalable based on the demand of the end user. You want to distribute incoming network traffic across a group of back-end resources or services such as virtual machines (VMs). You must scale your applications while maintaining throughput and keeping response times low.

Solution

For scalability Azure provides you:

  • Azure Load Balancer. It works in Layer 4. You can manage VM, NAT, and the virtual network
  • Application Gateway. It works in Layer 7. You can manage HTTP, HTTPS traffic

On this case, we prefer to use Azure Load Balancer (since it's a VM)

  • Think the scenario, where you want to put the Azure Load Balancer
    • Public facing. For example, when you have a web server in a public subnet
    • Private facing. For example, when you want to manage load in private subnet such as database server
  • Select proper SKU
    • Basic – communicate with HTTP and TCP protocol, up to 300 instances
    • Standard – Basic SKU plus HTTPs and with SLA 99.99%
  • Choose the backend pool. Backend pool is containing the IP address or VM sets that already exist. For Basic you can only access VM in single network, while standard can access in availability set
    • Availability set
    • Single VM
    • VM Scale set
  • Configure rule of load balancer
  • Configure session persistence. It will help the user can use same VM when doing transaction
  • Create health probe. It rules how probe will make use the load balancer activate when certain health conditions

Learn More

blog comments powered by Disqus

Topics Highlights

About @ridife

This blog will be dedicated to integrate a knowledge between academic and industry need in the Software Engineering, DevOps, Cloud Computing and Microsoft 365 platform. Enjoy this blog and let's get in touch in any social media.

Xbox

Month List

Visitor