VM states (running, stopped, deallocated) in Azure - Time & Space Complexity

Time Complexity: VM states (running, stopped, deallocated)

O(n)

Understanding Time Complexity

When managing virtual machines (VMs) in Azure, it is important to understand how the time to change VM states grows as you manage more VMs.

We want to know how the number of operations changes when starting, stopping, or deallocating multiple VMs.

Scenario Under Consideration

Analyze the time complexity of changing VM states in a loop.

for vm in vm_list:
    if vm.state == 'running':
        vm.stop()
    elif vm.state == 'stopped':
        vm.deallocate()
    elif vm.state == 'deallocated':
        vm.start()

# Each operation calls Azure API to change VM state
    

This code changes the state of each VM in a list by calling the appropriate Azure API.

Identify Repeating Operations

Identify the API calls, resource provisioning, data transfers that repeat.

• Primary operation: Azure API call to change VM state (start, stop, deallocate)
• How many times: Once per VM in the list

How Execution Grows With Input

Each VM requires one API call to change its state, so the total calls grow directly with the number of VMs.

Input Size (n)	Approx. API Calls/Operations
10	10
100	100
1000	1000

Pattern observation: The number of API calls increases one-to-one with the number of VMs.

Final Time Complexity

Time Complexity: O(n)

This means the time to change VM states grows linearly as you add more VMs.

Common Mistake

[X] Wrong: "Changing the state of multiple VMs happens instantly regardless of how many VMs there are."

[OK] Correct: Each VM requires a separate API call, so more VMs mean more calls and more time.

Interview Connect

Understanding how operations scale with the number of resources helps you design efficient cloud management scripts and shows you think about real-world system behavior.

Self-Check

"What if we changed the code to batch start or stop multiple VMs in a single API call? How would the time complexity change?"

Practice

1. What does the deallocated state mean for an Azure VM?

easy

A. The VM is off and resources are freed.

B. The VM is running and using resources.

C. The VM is off but still reserving resources.

D. The VM is paused but still running.

Solution

1. Step 1: Understand VM states

   Azure VMs can be running, stopped, or deallocated. Running means active, stopped means off but resources reserved, deallocated means off and resources freed.

2. Step 2: Identify deallocated meaning

   Deallocated means the VM is turned off and Azure releases the compute resources, so you don't pay for them.

3. Final Answer:

   The VM is off and resources are freed. -> Option A

4. Quick Check:

   Deallocated = off + freed resources [OK]

Hint: Deallocated means VM off and no resources reserved [OK]

Common Mistakes:

• Confusing stopped with deallocated
• Thinking deallocated still uses resources
• Assuming deallocated means paused

2. Which Azure CLI command correctly stops a VM and releases its resources?

easy

A. az vm pause --name MyVM --resource-group MyGroup

B. az vm stop --name MyVM --resource-group MyGroup

C. az vm shutdown --name MyVM --resource-group MyGroup

D. az vm deallocate --name MyVM --resource-group MyGroup

Solution

1. Step 1: Review Azure CLI commands for VM states

   To stop a VM but keep resources, use az vm stop. To stop and free resources, use az vm deallocate.

2. Step 2: Identify command that releases resources

   az vm deallocate stops the VM and releases compute resources, matching the question.

3. Final Answer:

   az vm deallocate --name MyVM --resource-group MyGroup -> Option D

4. Quick Check:

   Deallocate command frees resources [OK]

Hint: Use 'deallocate' to stop VM and free resources [OK]

Common Mistakes:

• Using 'stop' which keeps resources reserved
• Confusing 'shutdown' with deallocate
• Using non-existent 'pause' command

3. If an Azure VM is in the stopped state, what happens to billing for compute resources?

medium

A. Billing doubles due to stopped state overhead.

B. Billing stops immediately as VM is off.

C. Billing continues because resources are reserved.

D. Billing pauses but storage costs continue.

Solution

1. Step 1: Understand billing for stopped VMs

   When a VM is stopped (not deallocated), compute resources are still reserved, so billing continues for compute.

2. Step 2: Differentiate from deallocated state

   Only when VM is deallocated are compute resources freed and billing stops for compute.

3. Final Answer:

   Billing continues because resources are reserved. -> Option C

4. Quick Check:

   Stopped = billing continues [OK]

Hint: Stopped VMs still reserve resources and incur charges [OK]

Common Mistakes:

• Assuming billing stops when VM is stopped
• Confusing storage billing with compute billing
• Thinking stopped means deallocated

4. You tried to start a VM but got an error because it is in the deallocated state. What should you do to fix this?

medium

A. Run az vm start to allocate and start the VM.

B. Run az vm stop to release resources first.

C. Run az vm restart to reset the VM state.

D. Run az vm deallocate again to confirm.

Solution

1. Step 1: Understand VM start from deallocated

   A VM in deallocated state has no resources allocated. To use it, you must start it which allocates resources and powers it on.

2. Step 2: Identify correct command to allocate and start

   az vm start allocates resources and powers on the VM, fixing the error.

3. Final Answer:

   Run az vm start to allocate and start the VM. -> Option A

4. Quick Check:

   Start command allocates resources [OK]

Hint: Use 'start' to allocate and power on deallocated VM [OK]

Common Mistakes:

• Trying to stop or deallocate an already deallocated VM
• Using restart which requires VM to be running
• Not allocating resources before starting

5. You want to save costs by stopping your VM but keep its IP address and disk data intact. Which state should you use and why?

hard

A. Use running state but reduce VM size to save costs.

B. Use stopped state because it keeps resources reserved including IP and disks.

C. Use deallocated state because it frees all resources including IP address.

D. Use stopped state because it frees compute resources but keeps IP.

Solution

1. Step 1: Understand resource reservation in stopped state

   When a VM is stopped (not deallocated), compute resources are reserved, so the IP address and disk data remain intact.

2. Step 2: Compare with deallocated state

   Deallocated frees compute resources and may release dynamic IP addresses, so IP might change.

3. Final Answer:

   Use stopped state because it keeps resources reserved including IP and disks. -> Option B

4. Quick Check:

   Stopped = keep IP and disks reserved [OK]

Hint: Stopped keeps IP and disks; deallocated frees resources [OK]

Common Mistakes:

• Choosing deallocated and losing IP address
• Thinking stopped frees compute resources
• Trying to save costs by resizing while running