With Spot Virtual Machines, Microsoft provides Azure cloud customers a pricing model that offers Azure compute capacity at 90% the cost of pay-as-you-go. Formerly known as Low Priority VMs, Spot VMs are allocated from Azure’s excess compute capacity and enable workloads to run for significantly reduced costs.
Azure VM pricing models
Customers can purchase and use compute from Azure as Spot VMs, as well as with pay-as-you-go and reserved instance options. Each pricing model has benefits and drawbacks, with your workload requirements helping to determine how you’ll balance and leverage these offerings for optimal performance. You can use a combination of any or all of these options. Regardless of which pricing model you choose, the compute capacity you receive is the same.
Spot virtual machines allow customers to purchase VMs from a pool of unused spare capacity at a significantly lower price—up to 90% less—than pay-as-you-go. The lower cost comes with the provision that Spot VMs can be taken away with minimal warning if demand for capacity increases or instances are needed to service reserved instances or pay-as-you-go customers. While the risk of disruption can make Spot VMs challenging for mission critical, production workloads that can’t afford a service interruption, significant cost savings can be achieved by running many kinds of workloads, from stateless, non-production applications to big data, on Spot VMs.
No long-term commitment or upfront payments, you can increase or decrease capacity as you need it and pay (by the second) only for what you use. This can be more expensive than Azure’s other pricing options, but this option often gives customers the flexibility and availability needed for mission-critical, unpredictable workloads. Learn more about Azure pay-as-you-go instances.
Reserved virtual machines
Azure’s reserved virtual machine instances offer customers a way to purchase compute capacity at 72% less than they would with pay-as-you-go. The discount is afforded with a one or three year commitment to a virtual machine in a specific region. Predictable workloads and long-term applications that need to run 24/7 can be a good match for reserved VMs. Learn more about Azure Reserved VMs.
What’s the difference between Spot VMs and Low Priority VMs?
Now decommissioned, Low Priority VMs were offered at a fixed price. With Spot VMs, customers can bid on spare capacity, and set the price they are willing to pay for compute. Prices for Spot VMs fluctuate based on demand and availability, and vary based on the capacity for size or SKU in an Azure region. If there are spot VMs that meet requirements and are priced below your cost ceiling, your workload will run. If the cost of spot VMs exceeds the maximum, or if Azure needs the capacity back, the VMs will be terminated.
Spot VM Considerations
When considering Spot VMs for your workloads to minimize operating costs, it’s important to keep some key characteristics in mind:
- Azure doesn’t offer any SLA for Spot VMs
- Termination can occur with only a 30-second warning
- Spot VMs are terminated based on capacity availability and your configured price maximum
- Available capacity varies based on region, size, and time of day
How do Azure Spot VMs work?
When creating a Spot VM, users define the eviction policy and determine the price controls. The default eviction mode for Spot VMs is stopped-deallocated state, which allows the VM to be deployed later if compute capacity becomes available. You can change the eviction to policy to delete, which would permanently remove the VM.
VMs can be evicted based on capacity or the maximum price you set. When other customers request more compute, the price of Spot VMs will increase. With a max price set, your VMs will terminate if the cost exceeds the limit you set.
Running Spot VMs are then managed just like any other Windows or Linux VM, and functions just like pay-as-you or reserved instances.
To make Spot VMs a real part of your cloud cost optimization strategy, learn more about Spot by NetApp’s Azure solutions.