Many people are all too familiar with the term “buyer’s remorse.” It typically happens when a customer buys a big-ticket item that they aren’t sure they can afford. Initially caught up in the excitement of the transaction, they start regretting their decision when reality sets in. Buyer’s remorse isn’t something that just happens to consumers. In fact, you could say that many IT departments are experiencing a type of buyer’s remorse when it comes to their cloud migration projects.
According to the 2019 Enterprise Cloud Index, a survey by Nutanix and Vanson Bourne, nearly three-quarters of respondents reported migrating applications from the public cloud back to a private cloud. When IDC studied the issue, they found the top repatriation drivers to be security (19%), performance (14%), cost (12%), control (12%), and the reduction of shadow IT (11%).
But the results from both the Nutanix and the IDC study can be a bit misleading if taken at face value. For example, IDC found security to be the top repatriation driver, and only 9% of the respondents in the Nutanix survey said the public cloud was “the most secure” operating model. But is the public cloud really less secure?
Like a lot of buyer’s remorse situations, the problem doesn’t lie in the product or service itself, so much as it does in the organization’s decision making processes or IT maturity. For example, the public cloud need not be less secure than any other type of cloud. It’s only the organization’s inability to deploy and manage workloads securely that makes it so. The same can be said of other repatriation drivers, such as performance and cost.
2 examples of unforeseen cloud migration woes
What such high repatriation numbers suggest is that organizations aren’t doing enough due diligence before migrating workloads to the cloud. The cloud migration team is either over-estimating the cloud platform’s ability to handle an organization’s needs or over-estimating the organization’s ability to handle the cloud platform. Let’s take a look at a couple of recent examples from our own customer files to illustrate the point.
We’re paying how much for AWS?!
One of the most common reasons to move to the cloud is to reduce CapEx expenditures in favor of OpEx. But that advantage quickly dissolves if the organization doesn’t keep a close eye on their monthly cloud spend.
In a recent webinar, Delivering Successful Outcomes with Cloud, Nutanix’s senior solutions marketing manager, Kong Yang related the story of a company that was spending tens of millions of dollars every month on AWS. Even in a large organization, you’d think that kind of invoice would raise all sorts of red flags.
In this case, it went largely unnoticed because there wasn’t just one invoice. There were hundreds of smaller invoices, often for a few thousand dollars or less. The overarching costs weren’t readily apparent until all those individual expenditures were rolled up. So long as the individual budget managers didn’t overspend, no one questioned it.
This kind of spending issue combined with a lack of governance can be a particular problem with public cloud infrastructure like AWS. It’s easy to spin up instances in development or for temporary projects and forget to spin them back down when the resources are no longer needed. This organization didn’t have the protocols in place to manage their spend nor the tools in place to give management an overall picture of their spend.
To add insult to injury, the organization had decided not to discontinue its on-prem data center. When we analyzed their resource utilization, we found it to be roughly 2%. So, for all they were spending every month for AWS, they were still paying the high overhead of maintaining an on-prem data center and managing all of their existing infrastructure.
Our cloud environments are way too complex
Gartner predicted that, in 2020, 75% of organizations will deploy a multicloud or hybrid cloud approach to the cloud. Migrating applications to one cloud is a difficult task, managing multiple migrations, maintaining multiple clouds, and ensuring interoperability is even harder.
IT departments often find themselves asking:
- Can I manage another migration with a shrinking budget and staff?
- Does my staff have the expertise to successfully migrate and manage multiple clouds?
- What type of cloud environment is best for my workloads: public or private, software as a service, platform as a service or infrastructure as a service?
- How do I protect or backup my workloads once I migrate to the cloud?
A concerning trend shows that original cloud migrations are taking longer and cost more than expected when proper planning and expertise is missing. One of the biggest culprits is the lack if cloud expertise and experience. Experts say cloud platforms are and will continue to evolve rapidly.
We can help you with cloud migration due diligence
On average, an IT professional may see one or two data center or cloud migrations in their career. It's unrealistic to expect people to be experts in something they will see/experience so infrequently. A managed cloud provider might be able to help you with your cloud migration or repatriation efforts, and introduce you to new technologies, like Hyperconverged Infrastructure, to help you avoid some of those biggest pitfalls that come with adopting the cloud. Because we’ve managed hundreds of migration projects, chances are good we will spot pitfalls your internal team can’t see. If you’d like to learn more about our cloud migration services, visit us on the web or reach out to one of our cloud migration advisors.