Recently, Microsoft introduced a new commercial offer in Azure called ‘Saving Plan’. In this blog, we will look more closely what that is, how it can be applied and how it differs from ‘Reservations’.
Before we explore the new Saving Plans, let’s remind ourselves what Reservations, such as Azure Reserved Virtual Machine(VM) Instances are and how these work. It will help later on, to understand the differences that a Saving Plan offers.
A Reserved VM Instance is a financial commitment for 1 or 3 years for a specific instance size, family and region. Only if all 3 match, will the reservation discount apply.
With such a fixed commitment comes a great cost saving. Microsoft suggests that up to 72% cost saving can be achieved compared to pay-as-you-go (PAYG) pricing. Furthermore, if an existing Windows license can be applied via Azure Hybrid Use Benefit (AHUB), then the discount can be as much as 80% compared to PAYG pricing. There are huge variations on discounts depending on the VM family and size.
The Reserved VM Instances can be paid for in full at time of procurement or cost can be spread to monthly payments.
Should the requirements change and a Reserved VM Instance is no longer required, this can be cancelled or exchanged. An exchange can only be performed where the value of the new VM reservation is equal or greater to the refunded VM reservation.
Reserved VM Instances can be cancelled at any time (up to £46,526.77 per year). Cancelling allows you to return the remaining months of an Azure Reserved VM Instance purchase to Microsoft. Currently, there is no cancellation charge, but in the future, there will be a 12 percent early termination fee.
A good use case for a Reserved VM Instance is a server that is required all-the time, such as a Domain Controller or a SQL server VM (where there is a need for this to be a VM) or for a business application that requires availability for users or services 24×7.
Microsoft calculates a month as 730 hours and so a 3 year reserved instance means an uptime of such a Reserved VM Instance 24×7 for 36 months.
Depending on the VM size and family, the discount varies and with that the return-on-investment (ROI). Some 3-y Reserved VM Instances break even within 18 months.
So it is important to calculate the business case to ensure that the commercials stack up.
Azure Saving Plans
The latest commercial offering, the Azure Saving plans take a different approach.
Instead of a blanket commitment for 1 or 3 years for 24×7 hours, saving plans can be defined for a commitment for a fixed amount of compute hours regardless of type, family or size of the VM and regardless of where in Azure the VM is and can be assigned at enrolment or subscription level.
Should the actual usage exceed or go beyond the Compute saving plan, normal PAYG pricing will be applied. Multiple compute saving plans can be procured and always the maximum discount will be applied.
Compute Saving plans also don’t consider the OS and instead offer a blended cost. If a Windows license is required, customers can bring their own license via AHUB.
Different to Azure Reserved VM Instances, Saving plans cannot be cancelled and hence no refund or exchange could be achieved.
For a limited time, you can trade in your virtual machine, Azure Dedicated Host, and App Service reservations for an Azure savings plan for compute. The total value of the Azure savings plan for compute must be greater than or equal to the prorated remaining monetary value of the reservation commitments you want to trade in. The new Azure savings plan for compute starts a net new term from the trade-in date. You can trade in a recently purchased reserved instance for a savings plan in the Azure portal.
Summary
In both cases, Microsoft will provide recommendations via Azure Advisor if a saving can be achieved.
Because the way Saving Plans apply discounts, it needs careful consideration which commercial offering is best suited for the individual business case.
Would you like to know more?
As an Azure Expert MSP, risual have deep technical knowledge of Azure and are currently building these platforms for multiple clients right now! Contact us on the link below if your organisation wants to take advantage of a modern cloud technology approach to solving data problems.
