There are many ways to save money in the cloud, from rightsizing resources to renegotiating cloud service contracts. When implemented properly, all of these measures will reduce cloud spending to some degree.
However, some cloud savings strategies tend to yield higher results than others – especially if you measure benefits in terms of money saved versus the time and effort you have to invest in planning and implementing cloud cost optimization methods.
With that reality in mind, let's explore which cloud cost optimization strategies deliver the biggest bang for your buck, as well as those that yield poorer results relative to what they cost in time and effort.
The most efficient ways to reduce cloud spending
To achieve the greatest possible cloud cost savings while putting in the lowest effort, you should look for opportunities to slash cloud spending across the board, rather than reducing the costs of individual cloud resources.
Here are a few ways to do that.
1. Discounted cloud pricing agreements
The most obvious way to do this is to negotiate discounted cloud service pricing. In most cases, taking advantage of this option requires you to consume a significant amount of cloud services, because cloud providers typically won't discount pricing for small-scale customers. But if you spend at least several hundred thousand dollars a year on cloud services, a negotiated discount pricing agreement might be feasible, especially if you're willing to commit to a multi-year service contract.
The benefit of discounted pricing agreements is that they are a simple way of reducing costs on potentially every cloud resource that you deploy. You don't have to evaluate the cost effectiveness of individual workloads or track workloads to identify waste. Instead, in exchange for the upfront effort of negotiating a service agreement, you get discounted pricing on an extended basis.
2. Startup credits
What if your organization is smaller and doesn't spend enough in the cloud to get cloud providers to negotiate pricing discounts for you? In that case, look into startup credits that your cloud provider offers. All of the major public clouds (not to mention some colocation companies) offer credits designed to provide free cloud services to startups.
There is a limit on how many startup credits you can get; in most cases, don't expect more than low-six-figure packages. Cloud providers may also not offer startup credits to businesses that don't have a lot of growth potential, because those businesses are not likely to become major paying customers down the line. But in cases where you can get them, startup credits are another excellent way to reduce cloud spending with minimal time and effort.
3. Discounted cloud services
A third way to save substantial money in the cloud with relatively little investment of time and effort is to take advantage of discounted cloud service offerings. For example, you can use reserved VM instances to reduce your spending on cloud VM hosting.
Not all cloud services are available at discounted costs, and not all workloads are compatible with discounted instance types. But for those that are, this is an easy way to save money without having to modify workloads or carefully track spending.
How not to save money in the cloud
Now, let's look at some examples of strategies that will typically reduce your cloud spending, but not in a time-efficient way.
1. VM instance rightsizing
VM instance rightsizing is the process of ensuring that your VM types are ideally suited for the workloads they host.
You certainly should strive to rightsize – or, better, choose the right instance types out of the gate, so that you don't need to reconfigure them later. But rightsizing takes a lot of time and effort because it requires you to assess individual workloads and pore over the hundreds of VM instance types that could potentially host them. And when you do rightsize a VM instance, you save money only on that particular instance, which might represent a tiny fraction of your total cloud computing bill.
2. Cloud service migration
Migrating workloads to different types of cloud services may save you money. For example, it might be more cost-effective to run a certain workload using a serverless computing platform than VMs.
The downside of this strategy, however, is that migrating workloads often requires major reconfiguration or, in some cases, refactoring the code of your applications themselves. And as with VM rightsizing, you only save money on individual workloads. It's a high-effort activity that yields relatively little cost savings.
3. Multicloud
Cost savings are often touted as one of the main reasons to adopt a multicloud architecture. The idea is that when you use multiple clouds, you can pick and choose between more cloud service offerings to get a better balance between pricing and performance.
While that's true, the downside of multicloud is that it can substantially increase your administrative overhead. You have to manage two or more separate cloud environments and master the tooling that is built into each one.
Conclusion
For the record, let me state clearly that I'm all for cloud cost optimization strategies like workload rightsizing or adopting a multicloud architecture. There's nothing wrong with those methods, and embracing them certainly makes sense if you've already pursued more efficient means of reducing your cloud bill.
But if you haven't, start your cloud cost optimization strategy by focusing on the low-hanging fruit, such as discounted pricing agreements and service offerings. That's where the easiest cost-savings opportunities lie.
