Optimize the benefits of Azure savings plan for compute
Microsoft has been continually augmenting its portfolio of commercial offerings to benefit customers who, like your company, remain committed to using Azure services. Azure savings plan for compute supplements that portfolio and represents an extra cost-saving option available to existing and new customers.
Use multiple Azure cost-saving options
In general, compute cost comprises software and infrastructure components. To minimize the software cost of Windows Server and SQL Server workloads, Azure Hybrid Benefit-eligible customers can use existing on-premises licenses. The optimal approach to minimizing compute charges depends on your workloads:
- Azure reservations offer the maximum cost savings for stable, predictable workloads running on specific virtual machines in a specific Azure region.
- Spot virtual machines are optimal for interruptible workloads.
- Azure savings plan for compute is most suitable for customers who are looking for a flexible way to reduce costs from on-demand compute usage.
Some of the more common use cases for Azure savings plan for compute include scenarios where you have:
- Centralized budgeting and procurement.
- Dynamic workloads that move across data center regions (for example, in the follow-the-sun scenarios).
- An actively evolving IT roadmap (for example, a planned transition from the Infrastructure-as-a-Service to Platform-as-a-Service model).
- A need for automating cost optimization across various services.
You can use both Azure savings plan for compute and reservations. Microsoft automatically maximizes your benefit and savings across eligible usage each hour. Microsoft always applies the reservations first due to their more targeted scope and higher savings rates. Azure savings plan then applies to the remaining eligible usage until the commitment is fully consumed. After a reservation or an Azure savings plan for compute is fully consumed or expires, you're billed on a pay-as-you-go basis.
Let’s assume that your compute cost totals $1,000,000 per year billed on a pay-as-you-go basis. By allocating costs to the appropriate savings options based on your needs, you can optimize the cloud budget usage in the following manner:
- Purchase reservations for the most stable resource needs, lowering the $300,000 costs to $150,000.
- Purchase an Azure savings plan to cover dynamic resource needs, lowering the $500,000 costs to $200,000.
- Continue paying only for what you use for the most inconsistent resource needs with no commitment required.
Individual savings options are subject to their respective terms based on their purchase dates.
To realize optimal savings, select a longer term commitment and ensure the benefit scope has sufficient usage and hourly commitment is equal or lesser than the recommended amount. You can determine actual savings by comparing the lower savings plan prices vs pay-as-you-go prices. You can review personalized recommendations based on the recent on-demand usage of eligible resources. Recommendations are shown in the Azure portal, Azure Advisor, and through the Recommendations API. To summarize, a savings plan is designed for users who have consistent spend on compute resources, but need flexibility to accommodate changes. Reservations are a great option if you know exactly what you need and are looking for more savings.
In your case, you can continue using your existing reservation commitments, which cover services such as Cosmos DB and Azure Storage. For others, you should consider the suitability of Azure savings plan for compute to workloads that transition across Azure regions in a follow-the-sun manner or if you plan to rearchitect them to take full advantage of the cloud capabilities.