How to calculate cloud migration costs before you move

4 min read

Determining the cost of a cloud migration isn’t easy. Not only do businesses need to account for the differences between on-premises and cloud prices, they also must consider a variety of other factors — many of which are often overlooked.

Some costs are easy to predict. For example, the cost of migrating data from on-premises storage to cloud-based object storage is straightforward. Other migration costs, like those incurred from workload refactoring, are more difficult to nail down. It’s also easy to overlook expenses associated with things like staffing and deploying new types of services.

Factor in the following costs to make sure a cloud migration will be financially beneficial.

The first step to calculate cloud migration costs begins before you move anything to the cloud. Admins need to assess the cost of existing hardware and software assets, then evaluate how it compares to a cloud-based environment. The challenge in this comparison is that cost models for most on-premises software and hardware are different from cloud pricing models.

On-premises computing requires a significant upfront capital investment to purchase hardware and follows a Capex-based cost model. On the other hand, cloud resources rarely require capital expenditure and follow an Opex-based model.

Customers pay for IaaS-based virtual hardware and SaaS applications as they consume them. This means you’ll need to express your on-premises Capex cost in a way that allows you to compare it to Opex costs in the cloud.

To do so, divide the upfront cost of your on-premises resources by the amount of time you can reasonably expect to use them. These estimates aren’t exact. They don’t account for costs such as replacing the server’s hard disks, which may not last as long as the server itself. It also does not factor in the potential for hardware upgrades, like the addition of memory, that could extend the server’s lifecycle. Still, this approach helps you establish a baseline estimate of the total cost of your on-premises environment, then compare it to the cost of equivalent services in the cloud. You also need to identify on-premises resources that you won’t pay for in the cloud.

For example, network switches for your on-premises data center are unnecessary when you move workloads to a public cloud. Uninterruptible power supply units and network-attached storage devices are also examples of equipment that you can decommission when you move to the cloud. Certain on-premises operating expenses, such as electricity and physical site security, also disappear post-migration.

In the simplest scenario, admins will take applications that currently run in on-premises VMs, as well as data housed in scale-out on-premises storage, and move them to public cloud compute and storage services. In this case, workloads won’t require any major refactoring, and the cloud services will have relatively simple pricing models. Migration costs are easy enough to calculate.

On the other hand, your cloud migration plan might not be to simply lift and shift your workloads into the cloud, but may also entail the transformation of those workloads. In these cases, cloud migration will require more development effort to modify your workloads. For example, you might run your applications inside VMs but want to migrate some to containers and serverless functions. Or, you have monolithic applications that you plan to refactor as microservices. These modifications can be expensive. Using a more complex menu of cloud services often requires more expertise to administer effectively, which can also translate to more expenses.

After you estimate the cost of your on-premises environment, you can calculate the costs of the cloud environment you plan to build and compare the two. Virtually all your cloud spending will be on operating expenses that are billed monthly. However, calculating cloud costs is difficult because there are so many variables.

Continue Reading

Enjoyed this summary? Read the complete article at the source:

Continue at searchcloudcomputing.techtarget.com →

Yves Mulkers

Yves Mulkers is the founder of 7wData and a widely followed voice in the data and AI community. He curates the 7wData and AI Beat newsletters, reaching hundreds of thousands of data and AI professionals, and writes on data strategy, analytics, AI, and the evolving data ecosystem.