New See exactly what you're overpaying AWS in under 60 seconds. Try the Calculator for free →
Get a Demo

Azure Spot VMs

Azure Spot VMs are virtual machines that run on unused Azure capacity at significantly reduced prices, but can be evicted by Microsoft with 30 seconds notice when that capacity is needed elsewhere.

How It Works

When Azure has unused compute capacity in a region, it makes that capacity available as Spot VMs at a discount off the standard pay-as-you-go price. You set a maximum price you are willing to pay per hour. If the current Spot price stays below your maximum, your VM runs. If Azure needs the capacity back, or if the market price exceeds your maximum, Azure sends an eviction notice and terminates the VM within 30 seconds. Because of this, Spot VMs are suited to workloads that can tolerate interruption: batch processing, rendering, CI/CD pipelines, stateless web servers, and data analytics jobs that can checkpoint and resume. On AWS, the equivalent is Spot Instances. On GCP, the equivalent is Preemptible VMs or Spot VMs.

Why It Matters for Cloud Cost

Spot VMs can meaningfully reduce compute costs for the right workloads. The discount off on-demand pricing varies by VM size, region, and current demand. The risk is eviction: if your architecture is not designed to handle sudden termination, Spot VMs introduce reliability problems that outweigh the savings. Teams that over-rely on Spot without fault-tolerant design can face failed jobs, data loss, or service disruptions. For predictable, always-on workloads, Azure Reservations or Azure Savings Plans are a better fit because they provide guaranteed capacity with no eviction risk. Spot VMs are best used as a cost layer on top of a committed baseline, not as a replacement for it.

For predictable Azure workloads that are a poor fit for Spot, Usage AI’s CoPilot surfaces projected savings from Azure Savings Plans commitments for customer review before any purchase is executed.

See how Usage AI saves 30 to 50% on AWS, GCP, and Azure.

Related Terms

Amazon EC2

Amazon EC2 (Elastic Compute Cloud) is AWS's on-demand virtual server service, providing resizable compute capacity in the cloud and representing the largest share of AWS spend for most companies.

AWS Reserved Instances

AWS Reserved Instances (RIs) are a commitment-based pricing model that lets you reserve capacity for AWS services in exchange for discounts of up to 72% compared to on-demand rates.

Azure Reservations

Azure Reservations are pre-purchase commitments for Azure resources that reduce costs by up to 72% compared to on-demand pricing in exchange for a 1-year or 3-year term.