Amazon Relational Database Service (RDS) makes it easier for organizations to set up, operate, and scale relational databases in the cloud. However, many AWS users soon discover that just like with Amazon EC2, RDS costs can be difficult to manage, especially as operations scale.
While identifying and optimizing resource utilization (e.g. downsizing) or stopping and starting your RDS instances on a schedule are both excellent ways to reduce your monthly AWS bill, purchasing Reserved Instances at a one- or three-year term in return for massive discounts off on-demand pricing is another great option for organizations who have insight into their long-term database usage plans.
Amazon RDS Reserved Instance options
Amazon RDS Reserved Instances give you the option to reserve a database instance for either a one- or a three-year term in exchange for a significant discount compared to on-demand instance pricing for the database instance.
Similar to the Reserved Instance offerings you’ll find for EC2, AWS currently offers three RDS Reserved Instance payment options:
- No Upfront: Offers a significant discount (~30%) compared to on-demand prices and is available with a one-year term
- Partial Upfront: Offers a higher discount than No Upfront Reserved Instances (~60% for a three-year term) and balances the Reserved Instance payments between upfront and hourly
- All Upfront: Offers the higher discount of all the Reserved Instance payment options (~63% for a three-year term) and is available with a one- or three-year term
One thing to note is that you can’t cancel an RDS Reserved Instance, but if you delete an RDS instance that’s covered, you can launch another instance with compatible specifications. In this case, you continue to get the discounted rate during the reservation term. Additionally, unlike EC2, RDS doesn’t currently support convertible reservations.
To learn more about AWS RDS Reserved Instances, including information on billing, size flexibility, steps on how to purchase, see AWS’ documentation.
Managing AWS RDS Reserved Instances with CloudHealth
The decision to make an accurate and effective Reserved Instance purchase can be a daunting, resource-intensive task. To help with this, the CloudHealth RDS RI Optimizer evaluates your RDS instances to determine if you can benefit from the use of Reserved Instances, and also analyzes your previous usage data to identify the optimal purchase for your given budget.
When creating a new quote with CloudHealth’s RDS RI Optimizer, you can define settings based on what is most ideal for your cloud environment. This will ensure your quote provides the most accurate recommendations.
Some of these custom settings include:
- Time Period: Specify the time period over which you want your on-demand usage to be analyzed. Select a time period that is most reflective of the usage you expect in the future. Our best practice is to analyze usage over the past month, though a more static environment will likely benefit for analysis over a longer period of time.
- Preferred Term: While three-year terms provide the most cost savings, a safer better option might be to choose a one-year term to hedge against the potential for new instance types becoming available during the term.
- Estimated Purchase Date: This date is critical if you have existing reservations that will expire before the expected purchase. When you choose a date, the Optimizer takes expiring reservations into account, ensuring that any renewals included are subject to your budget constraints.
- Reservation Type: Be sure to include only the reservation types you want to be considered in your RDS RI Optimizer recommendations. For example, some organizations might prefer not to use the All Upfront payment option and should remove that from the quote ahead of time.
- Choose Lower Prepay Reservation Type if Savings < X%: This setting downgrades the payment option if the savings difference between reservation types is smaller than the specified percentage. For example, if the savings from purchasing Partial Upfront RIs over No Upfront RIs is <3%, the recommendation downgrades to suggest that you purchase No Upfront RIs, which have a lower prepay than Partial Upfront RIs.
After running an evaluation, the CloudHealth RDS RI Optimizer returns recommendations for purchasing or modifying Reserved Instances, such as the example you can see in the image below.
CloudHealth will show you your on-demand usage overlaid on the recommended Reserved Instance usage by type, helping you see what your usage looks like today and how we recommend this expense be managed with Reserved Instance purchases.
You can always modify these recommendations and add filters to ensure optimization aligns with your organization’s needs. For example, you might want to make a Reserved Instance purchase for specific accounts, regions (e.g., US East Region), instance types (e.g., db.m3 instance types), functions (e.g., Elasticsearch Clusters), or even business groups (e.g., marketing department).
Before committing to the order, easily visualize what your infrastructure will look like if you make purchases based on the recommendations, and share this information with key stakeholders to gain agreement. Then, simply purchase your Reserved Instances through the CloudHealth Platform, your Amazon account representative, or the AWS Console.
Managing your Amazon RDS costs is crucial for long-term financial success, but to do so properly requires resource utilization monitoring coupled with complex Reserved Instance analyses—both of which can be time-consuming and prone to error in their own right. With the right modeling tools in place, you can confidently make discount purchases to help reduce your organization’s bottom line.
For more information about CloudHealth's Reserved Instance management capabilities, see our solution brief here. And to see how Reserved Instances fit into your broader cloud financial management strategy, see our in-depth whitepaper: Building a Successful Cloud Financial Management Practice