5 Cloud Cost Optimization Best Practices You Might Have Missed

If you help to manage cloud resources, you can probably rattle off a list of core best practices for keeping cloud spending in check. You likely know about the importance of selecting the right VM instance type for each workload, for example, and of taking advantage of reserved instances where feasible.

I don't want to write about practices like these. Although they're effective, they're not especially original or novel. Most businesses that want to save money in the cloud are already doing these things.

Instead, I'll look at cloud cost optimization strategies that are easy to overlook. Once you've implemented basic cloud cost management practices, there may be additional steps you can take to reduce your spending even further without compromising on workload performance.

Keep reading for a look at five practices that can substantially reduce your cloud spending, but that many organizations fail to leverage.

1. Optimize Spending in Non-production Cloud Environments

When businesses embark on cloud cost optimization programs, they typically start with production workloads, meaning those that power core business operations. The cloud resources that power production workloads often account for the majority of cloud spending, so it makes sense to reduce costs in production environments first.

Related:5 Myths About Cloud Pricing

But many organizations also have cloud environments where they host applications that are under development or in testing. Those environments, too, are important places to implement cost-reduction strategies.

What's more, it's important to understand that the cost optimization practices you've implemented for production may not be ideal for your development and testing environments. For example, in a production environment you might decide to retain cloud snapshots (which can help back up VM images) for a 30-day period, whereas in a development environment you only require a seven-day retention policy to meet your backup and recovery goals. In this case, you'd be wasting money by storing snapshots for development workloads for the same period of time as for production workloads.

2. Use Intelligent Cloud Storage Tiering

You may know that cloud object storage services like Amazon S3 offer dirt-cheap storage tiers, where the tradeoff for low storage fees is that you can't always access your data right away. You may already be using those tiers to host data that you don't need to access frequently, such as logs that you have already processed but need to retain for compliance reasons.

That's a great way to save money when you have data that you know from the start you won't need to access frequently. But what about other data that you think you will need to access more frequently, but which turns out not to require regular access?

Related:6 Tips for Controlling Your Cloud Costs in a Recession

You can reduce the costs of storing that data by taking advantage of intelligent cloud storage tiering, an oft-overlooked feature of major cloud platforms. Intelligent storage tiering automatically monitors your data usage patterns and moves infrequently accessed data to lower-cost storage tiers for you.

The takeaway here is that you don't need to identify upfront all of the data that should live in low-cost storage. You should also take advantage of intelligent storage tiering to cut your cloud storage costs down to the absolute minimum.

3. Leverage Application-Level Tagging

Using tags to label cloud resources can help reduce costs by making it easier for organizations to keep track of resources. Tags also help identify idle or underutilized resources. Most teams know these things, and most organizations that use the cloud at scale have at least basic tagging policies in place that require engineers to tag infrastructure resources that they deploy in the cloud, like VMs and databases.

What many teams overlook, however, is creating tags that identify the applications associated with a given workload, not just infrastructure. Application-centric tags are important because they help organizations track which departments or employees use each application, not just the infrastructure it's deployed on.

Plus, if you have a multicloud strategy, you might deploy the same application on multiple clouds. In that case, application tags are very valuable because they help you track application performance and cost across clouds so you can ensure you're getting the most bang for your buck, regardless of where you host the application.

4. Use Interim Cost-Savings Measures

It sometimes happens that organizations identify ways to save money in the cloud, but implementing the necessary changes will take time. For example, perhaps your cloud provider has introduced a new family of VM instance types that will reduce your spending, but migrating all of your workloads to the new instances will take many months.

In a scenario like this, it can be easy to overlook interim cost-savings measures that might not save you as much money, but that you can implement sooner. For example, you might take advantage of reserved instances to reduce your spending in the short term, while you also work toward migrating to the new instance types in the long run.

In other words, you should never treat cloud cost optimization as an all-or-nothing affair. Interim savings measures can have a notable impact on your budget, even if they fall short of delivering the savings available through other, more complicated initiatives.

5. Stay Up-to-Date on Cloud Instance Types

Speaking of cloud vendors releasing new instance types, that's a practice that happens frequently on most public clouds. And when clouds offer new instance families, their customers can often reap new cost-savings opportunities.

For example, you might be running your workloads today on EC2 instances that use Intel processors, but as Amazon continues to expand its lineup of Graviton-based instances, you might find that you can achieve the same performance at a lower cost by taking advantage of Graviton instances.

This is why it's important to monitor instance offerings continuously and evaluate whether the instances you initially selected are still the best fit for your needs.

Conclusion

Following basic cloud cost optimization best practices is the first step toward eliminating unnecessary spending in the cloud. You might reduce your costs by 5% or 6%, for example.

But if you really want to strip your spending down to the bare minimum — if you want to get into double-digit cost reduction — don't settle for the basics. Think outside the box and identify strategies that allow you to double-down on savings while achieving the same, if not better, levels of workload performance.

Willy Sennott is the EVP of FinOps at Vega Cloud. He has 25+ years' experience in financial, marketing, and business analytics data, helping clients and companies drive revenue growth, improve cost efficiencies, and effectively allocate capital. At Vega Cloud, Sennott leads the FinOps practice and helps drive overall company strategy and product roadmap.