In addition to flexibility, cost is often a major factor for many cloud migrations. If you have time on your side, then it is a good practice to run a few apps in the cloud to understand the fine print and the value of cloud economics.
However, when you have to suddenly transition to the cloud in the face of a pandemic, time is not a friend; when you moved all the apps that you could, it was not entirely clear what the ensuing price tag would be.
Is That My Bill?
Today was the day you got your first bill. Boy, was that a shocker!
Yes, you can model the pricing and compare, but cloud providers have wildly varying ways of measuring and charging your application traffic and up-time, and the only way to really know what it’ll cost you is to use your applications in the real world.
Now that you have the bill, it is in your best interest to compare. There are many models to choose from – perpetual pricing per instance, bring-your-own license (BYOL), consumption and metered licensing models, by CPU cores, per-user, by throughput, and service provider-licensing agreements (SPLA), to name a few. What you choose eventually will depend on your need:
- Predictability in cost for you and your tenants
- Margins that you need to manage as a service provider or the ability to pass costs along to your tenants
- Application requirements to span both private and public cloud infrastructure
- Application profile in terms of resource consumption – are they changing or fixed?
How Do I Manage My Costs?
In a development environment (CI/CD) where services are added or updated frequently, support is required across multiple environments and across many stakeholders. So how do you manage your cost in such an environment?
- Know Your Apps: If your applications consume the same resources all the time, they may better suit traditional hosting, unlike a transient application that may benefit from pay as you go or metered pricing models.
- Know Your Cloud Provider: Take advantage of discounting mechanisms (spot pricing, reserved instances, commitment, etc.) to help you manage your cost depending on your application needs.
- Know Your Limitations – Automate: Enable usage alerts and notifications and explore auto scale capabilities to bring instances up or down depending on demand.
- Compare Your Cloud Provider Pricing vs. Other Elastic models: Many customers initially deploy on public clouds, but cost unpredictability becomes an issue once the services scale with usage. Flexible licensing is one of the critical elements to consider. Cloud providers are in it to make money and make cost comparisons difficult by design. Sometimes you have to watch out for changes in pricing. It helps to explore other pricing models to help you control costs depending on your application needs, such as license reuse across platform, costs predictability by throughput or on types of connections (SSL) that your applications may require.
- Analyze Data: It helps to profile and understand your application consumption to plan capacity and tie your applications to take advantage of discounting options with your cloud provider.
As organizations embark on a cloud journey, understanding the profile of their application infrastructure helps in managing cost. It is imperative to understand and compare with competing pricing models and take advantage of the cloud providers’ discounting mechanisms.