What Is ASG in Networking?

Imagine you are moving into a new apartment building with your friends. You all want to live close together, but you also want to be safe in case something goes wrong. If you all lived in the same room and the lights went out, all of you would be in the dark. But if you spread out across different floors and different parts of the building, even if one floor loses power, the others stay bright.

An Availability Set works exactly like that, but for virtual machines in the Azure cloud. When you create several virtual machines, which are like imaginary computers running inside Microsoft's datacenters, you can group them into an Availability Set. This set ensures that your VMs are placed on different physical servers, different racks of servers, and different power and cooling zones inside the datacenter.

Why does this matter? Because datacenters are huge buildings with thousands of servers. Sometimes a server breaks, or a network switch fails, or a power supply goes out. If all your VMs were on the same server, a single problem would knock them all offline. By placing VMs in an Availability Set, Azure guarantees that no single hardware failure can take down all your VMs at once. This is a basic but powerful way to make your applications more reliable without needing to build your own backup systems. For IT beginners, think of it as the cloud version of not putting all your eggs in one basket.

An Availability Set is an Azure resource that provides redundancy and high availability for virtual machines by distributing them across multiple fault domains and update domains within a single Azure datacenter region. When you create an Availability Set and then add VMs to it, Azure automatically assigns each VM to a specific fault domain and update domain to ensure that hardware failures and planned maintenance events do not affect all VMs simultaneously.

Fault domains represent a group of servers that share a common power source and network switch. In a standard Availability Set, Azure creates up to three fault domains. Each VM you add is placed into one of these fault domains, so that if a power failure or network issue affects one fault domain, VMs in the other fault domains remain operational. This is critical for maintaining service availability during unplanned outages.

Update domains are groups of servers that are updated during planned maintenance. Azure defines up to 20 update domains in a single Availability Set, though the default is five. When Microsoft performs maintenance to patch or upgrade the physical servers, it reboots servers in one update domain at a time, waiting for the previous domain to complete before moving on. This means that VMs in an Availability Set are never all rebooted at the same time during maintenance.

Technically, the Availability Set does not cost anything itself; you only pay for the VMs inside it. However, you must create the Availability Set before you create the VMs, and you must place the VMs in the same region and resource group. The VMs within an Availability Set should be identical in configuration to support the application load. For the AZ-104 exam, it is important to remember that an Availability Set provides resilience only within a single datacenter, not across regions. For region-level disaster recovery, you would need Azure Site Recovery or paired regions.

Windows and Linux VMs can both be placed in an Availability Set. There is no difference in behavior. Networking and storage constraints apply normally, but you should be aware that using managed disks is strongly recommended with Availability Sets, as unmanaged disks can introduce single points of failure. The Availability Set concept is one of the foundational high-availability patterns in Azure, along with Availability Zones and Virtual Machine Scale Sets.

In real-world IT, downtime costs money and damages reputation. Even a few minutes of unavailability can lead to lost sales, unhappy users, or compliance violations. Many IT professionals are responsible for keeping applications running 24/7. An Availability Set is one of the easiest and cheapest ways to protect against hardware failures in Azure.

Without an Availability Set, if you run multiple VMs for a single application, Azure could place them all on the same physical server. If that server fails or needs maintenance, all your VMs go down at once. This is a single point of failure that is completely avoidable. By using an Availability Set, you ensure that your VMs are spread across different hardware, so no single failure takes out your entire application.

Availability Sets are especially important for applications that are not designed to handle sudden failures on their own. For example, a legacy application that runs on two VMs with a shared database can benefit greatly. Even if the application itself is not built for horizontal scaling, using an Availability Set protects it from hardware-related downtime.

For IT professionals studying for the AZ-104 exam, understanding Availability Sets is essential because the exam tests your ability to design and implement high-availability solutions. You will be asked to choose between Availability Sets, Availability Zones, and Virtual Machine Scale Sets based on cost, complexity, and redundancy requirements. Grasping the basic concept of fault and update domains is a must for these questions.

You will encounter scenario-based questions where a company needs to ensure that their virtual machines remain available during both planned maintenance and unplanned hardware failures. For example: A company runs a pair of domain controllers in Azure. They want to protect against a single server failure and also against maintenance reboots. What should they use? The answer is an Availability Set with at least two VMs placed in it.

Another question type tests your knowledge of fault domains. The question might say: Your application runs on three Azure VMs in an Availability Set. A fire in one server rack destroys all servers in that rack. How many VMs will be affected? The correct answer is one, because each VM is in a different fault domain.

Configuration questions also appear. You might be asked to order the steps to create an Availability Set. The steps include: create the Availability Set in the desired region, specify the number of fault and update domains, and then create VMs inside that set during the VM creation wizard. Some questions will present the steps out of order and ask you to arrange them correctly.

Troubleshooting questions may involve a situation where VMs are not behaving as expected. For instance, an admin creates an Availability Set but then notices all VMs are in the same fault domain. This can happen if the VMs were created before the Availability Set or if the set was created with only one fault domain. The question would ask you to identify the issue and the correct resolution.

Finally, you might see comparison questions that list multiple high-availability features and ask which one offers protection within a single datacenter. The answer is Availability Set. Other options might include Load Balancer, Traffic Manager, or Azure Site Recovery, which serve different purposes.

In day-to-day IT work, understanding Availability Sets helps you design resilient systems without overspending on complex solutions. When you deploy applications in Azure, you always need to consider what happens when a physical server fails. An Availability Set is your first line of defense against such failures.

To implement an Availability Set correctly, you must create it first. Many beginners create VMs and then try to add them to an Availability Set later, which is not possible. You also need to decide on the number of fault domains. In most cases, the maximum of three is best because it gives you the most hardware diversity. However, if you only have two VMs, two fault domains are fine.

Professionals should also know that Availability Sets work best with managed disks. Managed disks ensure that the VHD files for your VMs are stored in different storage clusters, further reducing single points of failure. If you use unmanaged disks, you risk having all your VMs depend on the same storage account, which could fail.

What can go wrong? If you set only one fault domain, all VMs end up on the same physical rack. This is a common configuration mistake. Also, if you have more VMs than fault domains, multiple VMs will share the same fault domain, but that is acceptable as long as you still have distribution across update domains. Another issue is forgetting to add a load balancer. Without it, users cannot automatically reach a backup VM when the primary one fails.

In practice, Availability Sets are commonly used for domain controllers, file servers, and legacy applications that cannot or should not be scaled out automatically. They are also a prerequisite for certain SLA agreements from Microsoft. If you have at least two VMs in an Availability Set with all traffic served through a load balancer, you can achieve a 99.95% uptime SLA for the VMs.

For the AZ-104 exam, expect to see questions about configuring Availability Sets, choosing between options, and understanding fault and update domains. The key takeaway is that Availability Sets are simple, free, and effective for rack-level redundancy, but they are not a substitute for cross-region disaster recovery.