On paper, your cloud migration may seem simple enough. Cloud-migration planning typically boils down to figuring out which workloads will benefit from being moved to the cloud, which types of cloud services to move them to, and which method you will use to get the workloads into those cloud services—while also estimating what it's all going to cost.
In practice, however, effective cloud-migration planning requires much more effort and attention to nuanced details than you might think. A number of considerations can be all too easy to overlook—but doing so will cause your migration strategy to stall (or, worse, fail entirely).
Although these factors may feel daunting, they can be managed without leading into analysis paralysis. To provide guidance, this article looks at six common oversights in cloud-migration planning—with a focus on those that cause financial harm. It explains why the mistakes are dangerous and offers tips on avoiding them to maximize your chances of cloud-migration success.
1. Putting Too Much Faith in Cloud TCO Calculators
One of the first steps that businesses tend to make when considering a migration to the cloud is opening up a cloud-cost calculator and plugging in numbers to estimate what they'll spend in the cloud. This is followed by a series of refinements with vendors and internal stakeholders. The major public-cloud providers offer total-cost-of-ownership (TCO) calculators that you can use for this purpose, as do some third-party businesses that operate in the cloud ecosystem.
These calculators are useful for getting a basic ballpark estimate of what it will cost to run a particular workload in the cloud (assuming that the cloud services you want to price out are supported by the calculator, which is not always the case). From there, you can decide if the workload should move to the cloud.
The problem with cloud-cost calculators, however, is that they are blunt instruments. The ones designed by cloud providers themselves tend to underestimate total costs because they assume that your workloads will be optimally configured—and because the cloud providers have an interest in minimizing your predicted spending on their platforms.
In other cases, cloud-cost calculators—especially those developed by third parties that have their own business agendas to advance—may overstate cloud costs because they fail to factor in certain variables, such as how many workloads can be consolidated when migrating to the cloud or which percentage of workloads will remain on premises.
The point here is that, while cloud-cost calculators may be useful for very basic planning purposes, trusting them blindly is a huge mistake. You need to dig deeper to get real visibility into how much each workload will cost after it migrates to the cloud. You must also accept that there will always be some deviation between estimated cloud costs and actual cloud costs because you can never predict all spending variables with complete accuracy. TCO calculators can be a useful tool for migration-planning purposes—so long as you remember that they are far from perfect.
2. Ignoring Hidden Cloud Charges
In a similar vein, whether you use cloud-cost calculators or not, it can be easy to ignore or underestimate what you might call the cloud's hidden costs. These include things such as support charges, data-egress fees, fees charged for cloud-monitoring software, and so on.
On the surface, these fees may seem minimal because they are often much smaller, in absolute terms, than the cost of cloud-infrastructure resources. Egress costs amount to just pennies per gigabyte per month, for example—whereas a single EC2 instance might run you hundreds of times that amount.
Nonetheless, the hidden fees can add up quickly—especially if your migration plans don't involve optimizing workloads in a way that minimizes these costs (by, for example, reducing egress or shrinking the amount of data you ingest into cloud-monitoring software).
3. Failing to Assess Workload Context
When you plan a migration to the cloud, it can be tempting to treat all workloads inside your business as equal candidates for migration. This approach leaves you assessing each workload from a purely technical perspective as you decide whether its performance and cost requirements mean that you should move it to the cloud.
In reality, however, workloads are not just technical objects. They are inextricably tied to business context; you need to factor in this context when planning a migration. For example, one department inside your organization might be planning a rapid expansion—meaning that moving its workloads to the cloud (where they can continue to scale seamlessly) would likely be more beneficial than moving the workloads of a department that is shrinking.
The bottom line here is that it is critical to think about cloud-migration plans not just as a technical process but also as a business process. You must consider which parts of the business will benefit most from moving to the cloud—as well as what the timelines are for migrating different business units' workloads into the cloud.
4. Ignoring Cloud-Migration Speed Requirements
There are multiple ways to migrate workloads to the cloud. Some, such as the lift-and-shift method—which essentially means moving your workload from your premises to the cloud with few if any configuration changes to the workload—are fast and simple. Others, like refactoring—which involves making major changes to the workload so it runs more efficiently in the cloud—are complicated, time-consuming, and costly.
A well-crafted cloud-migration plan should reflect evaluation of which types of migration methods make sense for which workloads. Although it's not uncommon for organizations to want to refactor applications prior to migration, in many cases it makes more sense to prioritize speed and simplicity over perfection.
For instance, you might have a workload that is failing on premises due to lack of available infrastructure. Accordingly, moving it into the cloud as quickly as possible to restore stability would be more important than optimizing it prior to migration. You can always optimize later in this case; in the near term, your priority would be to get into the cloud to prevent availability disruptions.
5. Not Tagging Cloud Resources
A simple but critical tool for helping to manage cloud spending is tagging. All of the major cloud platforms allow their customers to label different cloud resources with tags that identify what the resources do. Tags are useful from a technical perspective, therefore, because they help to manage these resources.
Tags also allow businesses to track spending easily on a workload-by-workload basis, based on which tags are associated with each workload. As such, tags are important for financial-planning reasons, too.
You don't have to use tags (cloud resources will run perfectly fine if they are not tagged). It's wise, though, to establish a tagging strategy during the cloud-migration process because tags play a central role in cost management. Determine which tags you'll use and how you'll leverage tags within your cost-tracking workflows.
6. Not Planning Post-Migration Check-Ins
A final blind spot that appears in some cloud-migration plans is a lack of post-migration check-ins.
Cloud migration should never be a one-and-done affair. No matter how carefully you plan and execute migrations, there is always a chance that your workloads won't be optimized for cost and performance following the migration. Additionally, evolving business needs may dictate updates to workloads after migration has occurred.
For both of these reasons, it's important to schedule post-migration workload assessments. You don't want to wait many months or years to discover problems with cloud workloads; instead, look for them systematically and proactively via regular check-ins.
Getting More from Cloud-Migration Plans
It would be nice if cloud migration were as simple as deciding what to move into the cloud, then moving it. But it's not. To get the most value out of cloud migration—and to avoid unexpected surprises—do these things:
- Factor in a wide margin of error when estimating cloud costs.
- Identify all relevant cost factors in cloud migration.
- Know which parts of your business stand to benefit most from cloud migration—and what the opportunity costs of not moving particular business units' workloads to the cloud will be.
- Identify how quickly you need to move to the cloud to achieve maximum value.
- Tag cloud resources to maintain visibility into both cost and performance.
- Assess configurations post-migration on an ongoing basis.
Businesses must factor all of these nuances and easily overlooked variables into their cloud-migration plans—lest they shoot themselves in the foot down the line.
Keep learning
Choose the right ESM tool for your needs. Get up to speed with the our Buyer's Guide to Enterprise Service Management Tools
What will the next generation of enterprise service management tools look like? TechBeacon's Guide to Optimizing Enterprise Service Management offers the insights.
Discover more about IT Operations Monitoring with TechBeacon's Guide.
What's the best way to get your robotic process automation project off the ground? Find out how to choose the right tools—and the right project.
Ready to advance up the IT career ladder? TechBeacon's Careers Topic Center provides expert advice you need to prepare for your next move.
