Azure Load Balancer

Load Balancing

Load balancing involves distributing incoming requests across multiple servers or resources to optimize resource utilization, enhance throughput, reduce response times, and prevent any single resource from becoming overloaded.

At a very high level, we have the following solutions available on Azure:

1. Azure Load Balancer: This layer 4 (TCP, UDP) load balancer distributes incoming traffic among healthy service instances within virtual machine scale sets or individual virtual machines.
2. Azure Application Gateway: This layer 7 (HTTP, HTTPS) load balancer allows you to manage traffic to your web applications. It features SSL termination, cookie-based affinity, and Layer 7 request routing.
3. Azure Traffic Manager: This DNS-based load balancer directs traffic to the most suitable endpoint based on the chosen routing method and the health of the endpoints.
4. Azure Front Door is a global, scalable, and secure entry point for web applications, offering global HTTP(S) load balancing, DDoS protection, and CDN integration.

Azure Load Balancer

Azure Load Balancer enables you to distribute incoming network traffic across multiple resources, such as virtual machines within a virtual network, enhancing the availability and reliability of your application.

Referred to as ALB, it uses scheduling algorithms like round-robin or least connections to distribute traffic. For instance, it will direct the first incoming request to the first virtual machine, the second request to the second virtual machine, and so forth.

ALB will automatically redirect traffic to the remaining available virtual machines if one becomes unavailable. ALB can load balance services across multiple ports, IP addresses, or both.

Load balancers can be either public or internal. A public load balancer can handle outbound connections for VMs in your VNet and distribute client traffic from the internet across your VMs. In contrast, an internal load balancer uses private IP addresses as the front end to balance traffic from internal Azure resources to other resources within a VNet.

Azure Load Balancer offers two SKUs: Basic and Standard. The Basic load balancer is ideal for simple workloads, while the Standard load balancer is more feature-rich and suited for complex scenarios.

Azure enhances redundancy and scalability through availability zones, and physically separate data centers within an Azure region. Each availability zone comprises one or more data centers with independent power, networking, and cooling infrastructure. Using availability zones allows you to create highly available applications that can withstand the loss of a single data center.

Azure Load Balancer supports availability zone scenarios, and the Standard ALB can increase availability and distribute traffic across zones.

Azure services that are compatible with availability zones fall into three categories:

• Zonal services: A resource can be assigned to a specific zone. For example, virtual machines can be placed in a particular zone, enhancing resilience by distributing instances of resources across multiple zones.
• Zone-redundant services: These resources are automatically replicated or distributed across zones. Azure replicates data across three zones to ensure that a zone failure does not affect availability.
• Non-regional services: These are Azure services that are consistently accessible from Azure geographies and resilient to outages at both the zone and region levels.

The Standard SKU offers support for availability zones with zone-redundant and zonal frontends for both inbound and outbound traffic, along with health probes for TCP, HTTP, and HTTPS protocols.

In contrast, the Basic SKU only supports TCP and HTTP. Additionally, the Standard SKU provides the advantage of utilizing HA (High Availability) ports.

"In our upcoming post, we'll explore the fundamental functionalities of Azure Traffic Manager."

Microsoft Azure Microsoft Azure DevOps Microsoft Learn