Azure Spot VMs: Massive Savings for Interruptible Workloads

Azure Spot VMs offer up to 90% discount compared to pay-as-you-go pricing. The catch? They can be evicted when Azure needs the capacity back. Let’s explore how to use them effectively.

Understanding Spot VMs

Spot VMs use Azure’s spare capacity. When Azure needs the capacity for pay-as-you-go customers, Spot VMs receive a 30-second notice before eviction.

Creating a Spot VM

resource spotVM 'Microsoft.Compute/virtualMachines@2021-11-01' = {
  name: 'spot-vm'
  location: location
  properties: {
    hardwareProfile: {
      vmSize: 'Standard_D4s_v3'
    }
    priority: 'Spot'
    evictionPolicy: 'Deallocate'  // or 'Delete'
    billingProfile: {
      maxPrice: -1  // Current spot price, or set specific max
    }
    storageProfile: {
      imageReference: {
        publisher: 'Canonical'
        offer: '0001-com-ubuntu-server-focal'
        sku: '20_04-lts-gen2'
        version: 'latest'
      }
      osDisk: {
        createOption: 'FromImage'
        managedDisk: {
          storageAccountType: 'StandardSSD_LRS'
        }
      }
    }
    networkProfile: {
      networkInterfaces: [
        {
          id: nic.id
        }
      ]
    }
    osProfile: {
      computerName: 'spot-vm'
      adminUsername: adminUsername
      adminPassword: adminPassword
    }
  }
}

Spot VM Scale Sets

For higher availability, use Spot VM Scale Sets with multiple zones:

resource spotVMSS 'Microsoft.Compute/virtualMachineScaleSets@2021-11-01' = {
  name: 'spot-vmss'
  location: location
  sku: {
    name: 'Standard_D4s_v3'
    tier: 'Standard'
    capacity: 10
  }
  properties: {
    upgradePolicy: {
      mode: 'Rolling'
    }
    virtualMachineProfile: {
      priority: 'Spot'
      evictionPolicy: 'Delete'
      billingProfile: {
        maxPrice: 0.5  // Maximum price per hour
      }
      storageProfile: {
        imageReference: {
          publisher: 'Canonical'
          offer: '0001-com-ubuntu-server-focal'
          sku: '20_04-lts-gen2'
          version: 'latest'
        }
        osDisk: {
          createOption: 'FromImage'
          managedDisk: {
            storageAccountType: 'StandardSSD_LRS'
          }
        }
      }
      networkProfile: {
        networkInterfaceConfigurations: [
          {
            name: 'nic'
            properties: {
              primary: true
              ipConfigurations: [
                {
                  name: 'ipconfig'
                  properties: {
                    subnet: {
                      id: subnet.id
                    }
                  }
                }
              ]
            }
          }
        ]
      }
    }
    platformFaultDomainCount: 1
  }
  zones: ['1', '2', '3']
}

Handling Eviction Events

Set up eviction monitoring in your application:

import requests
import time
import signal
import sys

METADATA_URL = "http://169.254.169.254/metadata/scheduledevents?api-version=2020-07-01"
HEADERS = {"Metadata": "true"}

def check_for_eviction():
    try:
        response = requests.get(METADATA_URL, headers=HEADERS, timeout=5)
        events = response.json().get("Events", [])
        for event in events:
            if event.get("EventType") == "Preempt":
                return True, event
    except Exception as e:
        print(f"Error checking events: {e}")
    return False, None

def graceful_shutdown(signum, frame):
    print("Received shutdown signal, cleaning up...")
    # Save state, checkpoint, etc.
    sys.exit(0)

signal.signal(signal.SIGTERM, graceful_shutdown)

def main():
    while True:
        eviction, event = check_for_eviction()
        if eviction:
            print(f"Eviction notice received: {event}")
            # Start graceful shutdown
            graceful_shutdown(None, None)

        # Do actual work here
        do_work()
        time.sleep(5)

if __name__ == "__main__":
    main()

Best Practices

1. Checkpoint frequently - Save state so work isn’t lost
2. Use multiple regions - Spread workloads for availability
3. Set max price thoughtfully - Don’t set too high or you lose savings
4. Mix with regular VMs - Ensure critical capacity is always available
5. Use for stateless workloads - Batch processing, CI/CD agents, rendering

Ideal Use Cases

• Machine learning training
• Batch processing
• Dev/test environments
• CI/CD build agents
• Video rendering
• Big data analytics

Spot VMs are a powerful tool for cost optimization when you design your applications to handle interruptions gracefully.