The 4 Most Overlooked Azure Reservation Options for Cloud Cost Optimization

5 Min Read

Reservations, also referred to as Reserved Instances, are regarded as one of the most significant ways to save money in the public cloud. Microsoft introduced Azure Reserved Virtual Machine (VM) Instances in 2017, and has since evolved and expanded their reservation options to provide customers with greater flexibility and cost savings. 

However, many Azure reservation options tend to be overlooked or go unnoticed—leaving potential cost savings unutilized. In this article, I’ll cover the four most overlooked Azure reservation options you should consider to help lower your cloud bill. 

Azure SQL Database

Released in 2010, Microsoft’s Azure SQL Database server is a fully managed (PaaS) database engine that only continues to grow in popularity. In fact, according to db-engine.com’s rankings, it has jumped up 10 spots in the last calendar year, eclipsing other relational database models such as SAP HANA, Google BigQuery, and Amazon Redshift. 

To help offset costs, Microsoft offers a Reserved Capacity for Azure SQL Database and SQL Managed Instances. Just as with traditional Reserved Instances, reserved capacity allows users to make a one- or three-year commitment in exchange for discounts on the underlining compute costs of the service. You can pay for reserved capacity up front or monthly at the same price point, with no extra fees when you choose to pay monthly.

By utilizing reserved capacity, users can expect to save up to 33% on compute resources. Plus, if you already have SQL Server licensing, you can take advantage of the Azure Hybrid Benefit to save up to 80% collectively. One limitation to be aware of, however, is that you can only reserve vCore databases in SQL Database, not DTU.

You can learn more about optimizing Azure SQL Database for improved costs, security, and performance in our in-depth article here: How to Optimize Azure SQL Database

Azure Cosmos DB

If you have any experience with Azure Cosmos DB, you’ll understand just how quickly the bill for this service can add up. Released in 2017, Cosmos DB is a schema-agnostic, fully managed NoSQL database service for modern application development. The pricing model for Cosmos DB is a bit different than that of Azure SQL Database, and that’s also reflected in the way reserved capacity is handled for Cosmos DB. 

Cosmos DB has two main components that make up its pricing: storage (in GB/month), and Request Units per second (RU/s), which can be seen as its throughput. 

With Cosmos DB reserved capacity, the discount is applied on the throughput provisioned for Cosmos DB resources—not the storage or networking charges. Just as with other reservations, you can choose a one- or three-year commitment, and can expect to save up to 65% on regular prices.

Solution Brief
Master Reserved Instance Management with CloudHealth

Azure storage reserved capacity

Customers can also benefit from reserved capacity discounts on their Azure storage costs. As of today, the reservation benefit only applies to Blob Storage (GPv2) and Azure Data Lake Storage (Gen2). Users can expect to save up to 38% on storage costs for these services by purchasing reserved capacity for one or three years, and in increments of 100TB and 1PB sizes. What’s great as well, is that this is available for hot, cool, and archive storage tiers, and the discount is applied hourly on the total data stored within that hour.

For help unraveling the Azure storage pricing maze, see our article here.

Azure Synapse Analytics

If you haven’t heard of Azure Synapse Analytics, you may be more familiar with its predecessor, Azure SQL Data Warehouse. Simply put, Azure Synapse Analytics is the next evolution in Azure’s integrated analytics service offering. 

Like other Azure reservations, you commit to a reservation for a duration of one or three years, and you have the option to pay all upfront or with monthly payments. And similar to Cosmos DB’s capacity reservation, an Azure Synapse Analytics capacity reservation does not cover storage or networking costs associated with the service, but it does cover what’s known as the “cDWU” (compute Data Warehouse Units) that are consumed.

Managing Azure reservations

Here I’ve listed the four most overlooked Azure reservation options, but you can see a complete list of charges Azure reservations cover here, including the most popular: Azure Reserved VM Instances, which can save you up to 72% (and 82% when combined with Azure Hybrid Benefit) on your Azure cloud bill.

eBook
A Beginner’s Guide to Microsoft Azure Reserved Virtual Machine Instances

While the cost-saving benefits of Azure reservations are evident, identifying and managing reservations can be a manual and time-consuming process. Planning reservation purchases requires extensive analysis of your existing utilization and leaves you spending a lot of time in spreadsheets. Not only that, but post-purchase you’ll have to continuously monitor the resources to ensure a positive return on your investment and eliminate waste. CloudHealth takes the hassle out of Reserved Virtual Machine Instance management by enabling you to: 

  • Easily model out reservation purchases to understand the ROI and make the best decisions
  • Keep an eye on underutilized reservations you can exchange or cancel
  • Identify areas with high on-demand usage which indicate opportunities for additional savings with reservations
  • Stay on top of expiring reservations

You can learn more about getting started with CloudHealth reservation management in our solution brief here, or by reaching out to us directly here.

For more information on how to save costs on your Azure cloud bill, I recommend our eBook: 10 Best Practices for Reducing Spend in Azure

Rob McGuine, Technical Account Manager

Rob is a member of the CloudHealth Customer Success Team, focused on enabling customers to be successful in the cloud by providing insights into cloud management best practices, helping teams build a mature cloud strategy, and ensuring customers are maximizing their usage of the CloudHealth Platform. Prior to joining CloudHealth, Rob was a Customer Success Manager at Microsoft, where he was the main point of contact for Microsoft Azure customer engagements, driving Azure best practices, technical knowledge transfers, trainings, and other customer success initiatives.

We Think You Might Like These: