#24. Your Workload Optimizations Checklist — Part 1: Databases

Here is your comprehensive checklist to optimize Databases usage on Cloud

In this article we introduced the simple 80:20 rule that helps you to optimize workloads effectively and achieve cost savings through investigating the business value.

In this series of articles, we will consider together how to optimize the usage of services used by most organizations (i.e., compute, storage, databases, networking and SaaS), by following simple checklists that questions their business value. In this article we will start with Database Services.

The Database Optimisation Checklist

The following is a simple checklist for Database optimizations. It is divided into 5 reviews: business value, performance, technical, license and rates.

‍

Business Value Review

The first question is: Does this database still serve a business purpose? Many databases become obsolete over time—whether through unnecessary regional replicas or modernization efforts. A business value review can quickly identify opportunities to eliminate redundant database costs or even the entire database expense.

Another common cost driver is replication. While this practice is common and often necessary for resilience, replication is sometimes used for non-production instances. In such cases, a business review should flag and eliminate this redundancy.

Performance Review

Performance is related to system metrics such as memory, CPU, storage, and networking. A simple rule of thumb: if the database is running at low utilization—for example, less than 60% of CPU and memory—it's a good candidate for rightsizing (choosing a smaller size).

Storage medium also affects performance. Premium SSD volumes offer higher IOPS (Input/Output Operations Per Second) compared to standard HDD storage. This also applies to throughput (the amount of data read or written per second), latency (data access speed), and queue depth (the number of pending I/O requests a storage resource can handle simultaneously). While premium volumes are more expensive, most database operations don't require such high performance, making it logical to downgrade to standard storage.

Technical Review

Understanding which database type best suits your needs is crucial for optimization. Using the wrong database technology can lead to inefficiencies in both cost and performance. Ask yourself: "What purpose does this database serve?" Often, you'll find that your current database type may no longer align with your evolved business requirements. For example, a document database like MongoDB might be sufficient for use cases where a SQL database is currently being used.

This aligns with decisions about using cloud-native database platforms. Consider whether switching to services like DynamoDB (AWS), Cosmos DB (Azure), Firebase (GCP), or similar SaaS products would be beneficial. These platforms can drastically reduce costs and maintain high performance by shifting management responsibilities to the service provider.

When using SQL databases, consider database consolidation: This means combining multiple database instances into a single, more manageable, and efficient infrastructure. Consolidation not only simplifies instance management but can also reduce infrastructure and license costs while improving performance and security.

License Review

Are you using SQL? If so, review your licensing setup. For many use cases, a standard license is sufficient. Consider Enterprise licensing only when you need to meet specific business SLAs. If your database runs in a non-production environment, there's rarely justification for an enterprise license. Consider either downgrading your license or switching to open-source technologies (e.g., migrating from SQL Server to PostgreSQL).

For Azure users, it's worth examining Azure Hybrid Benefits (AHB). If AHB applies to your situation, you can avoid duplicate licensing fees.

Rate Review

Finally, review the rates for your database services. When using Cloud SQL services, you'll need a virtual machine to run the database. These VMs typically need to run continuously in the cloud, especially for production workloads. Instead of paying on-demand rates, consider using cloud commitment rates (such as RIs, SPs, or CUDs) which can save up to 72% on costs.

Summary

This article is part of a series that describes checklists for optimizing database, compute, and storage services. This one focuses on database optimization through reviews of business value, performance, technical aspects, licenses, and rates.