Subscribe to the Data Center Knowledge Newsletter
Get analysis and expert insight on the latest in data center business and technology delivered to your inbox daily.
5 Min Read
Image: Alamy
The cost of cloud services varies widely, but so does the functionality they offer. This means an expensive service may be well worth the price — if the capabilities it offers deliver a great deal of value.
On the other hand, some cloud services simply cost a lot without providing much in the way of value.
For IT organizations, then, a primary challenge in selecting cloud services is figuring out how much value they generate relative to their cost. This is rarely straightforward because what is valuable to one team might be of little use to another.
Nonetheless, there are concrete ways to assess whether a given cloud service is worth its cost for a particular team or use case. Read on for guidance as we unpack strategies for determining whether a cloud service offers true value.
The Challenges of Cloud Service Pricing and Value
In a simpler world, cloud service costs would be easy to predict, and teams would be able to easily determine whether a given service is creating real value.
In the real world, however, assessing the value of cloud services relative to their cost is deeply challenging for several reasons:
Pricing complexity: Most cloud services are subject to complex pricing arrangements that involve many variables — like how long you run a resource, how much data it ingests or stores, and how much data you move over the network. (This is why FinOps, the practice of assessing and managing cloud costs, has become an IT discipline in its own right.) This can make it hard to predict how much the service will cost you, let alone whether the price is worth it.
Varying features: The details surrounding cloud service features and capabilities are also very complex. Exactly what you can do with a given service may vary depending on which cloud region you use it in, for example, or (in the case of services like Amazon ECS or EKS) which "launch type" or "mode" you select.
Service changes: Cloud service providers update services on a frequent basis. In some cases, changes may mean that a service that once provided good value is no longer the best option for you. For instance, when Amazon introduces new EC2 instance types, they may turn out to be more cost-effective than ones you're currently using.
Changes to organizational needs: What your business needs from a cloud service can also change over time. For example, as you move more workloads into the cloud, you may find that cloud services offer more value because you can take greater advantage of integrations between your various workloads than you could if most of them ran on-prem.
Assessing the Value of Cloud Services
But just because the value of a cloud service is hard to assess and subject to change doesn't mean it's impossible to draw reasonable conclusions about whether a solution you're using is worth the price. Here are actionable ways to make a determination.
1. Consider whether the cloud service is essential
For starters, you can gain a lot of insight into the value of cloud services by identifying which ones are essential for keeping your business running. If you turned a particular service off tomorrow, would mission-critical applications or business processes stop running? If so, it's likely that that service is delivering real value (although it's still possible that you could make cost optimizations to increase its value further).
2. Compare pricing details across clouds
In some cases, one cloud vendor may offer the same type of service at a lower price, or with simpler pricing details. For instance, EKS, the managed Kubernetes service in the Amazon cloud, charges a fee for control plane operations, whereas AKS, the equivalent service on Azure, does not.
It's important, of course, to assess your total cloud costs, which are what ultimately matter most (it's not necessarily the case that AKS will always cost less than EKS, despite the free control plane). Still, if you're paying for something in one cloud that would be free in another, that could be a sign that you're not getting the greatest possible value.
3. Identify "hard" technical limitations
Cloud services may be subject to some specific limitations that you simply can't work around. For example, if you choose Graviton-based EC2 instances (which are typically more affordable relative to performance than x86-based instances), you won't be able to run x86-based code on them. That may be fine if you can recompile your code to support the Graviton architecture, but it is a major issue if you can't recompile — or if recompiling demands so much time and effort that it's not worth the cost savings.
In a case like this, it's important to assess how much investment you need to make to take full advantage of the cloud service, and whether it yields enough value in the long run to justify the upfront cost.
4. Anticipate pricing and feature changes over time
No one can predict how cloud service providers may change their pricing or features in the future, of course. But you can make reasonable predictions. For instance, there's an argument to be made (and I will make it) that as generative AI cloud services mature and AI adoption rates increase, cloud service providers will raise fees for AI services. Currently, most generative AI services appear to be operating at a steep financial loss — which is unsurprising because all of the GPUs powering AI services don't just pay for themselves. If cloud providers want to make money on genAI, they'll probably need to raise their rates sooner or later, potentially reducing the value that businesses leverage from generative AI.
That's just one example of a cloud service whose costs and value may change over time (and I could well be wrong in my predictions about AI pricing over time). But the point here is that if you adopt a cloud service at an early stage of its development or maturity, you should think about whether the service may change in the future in ways that impact how much value it delivers. Avoid becoming hooked on something that will become too expensive to justify in the future.
Read more about:
ITPro TodayAbout the Author
You May Also Like
