3 Common Application Migration Mistakes (and How to Avoid Them)
By: Ernest Sampera on August 9, 2021
The pressure to deliver cutting edge services to customers puts immense pressure on organizations to ensure that their IT stack is always in a position to deliver. Unfortunately, the substantial capital expense and technical toil that comes along with maintaining infrastructure often becomes too much of a burden to manage. For organizations that aren’t in a position to efficiently expand their on-premises footprint, the day eventually comes when they must consider carrying out an application migration. Due to data security and compliance requirements, many of them elect to move into a colocation data center.
However, even the simplest of application migrations can be quite complex undertakings. For companies that have utilized an on-premises IT environment for most of their existence, the transition to a third-party data center is a big moment that presents several opportunities for mistakes to be made. Here are a few of the most common miscalculations that can undermine a data center move.
Mistake #1: Not Understanding Dependencies
Whether an organization has been in business for five years or fifteen years, they’ve typically deployed a lot of applications across multiple systems and had data flowing between them in a variety of ways. In many cases, IT decisions are made quickly and many of the essential fixes and workarounds put in place over the years never make their way into documentation. The situation becomes even more opaque as knowledgeable personnel leave an organization or transition into different roles. When the time comesfor an application migration, IT managers may suddenly realize that they don’t have a clear map of their system’s dependencies.
It’s a problem that Jared Reimer, founder and President of Cascadeo, sees quite frequently during application migration projects. “Most companies that we’ve talked to do not actually understand all of the dependencies and data flows within their applications,” he says. “Even worse, they don’t have any proven ability to recreate their running system state.” If an IT department walks into a migration without a firm understanding of what will happen to connected systems when they move one application into another environment, data and business continuity could be seriously threatened.
Solution: Audit IT Systems
The very first step an organization should take before launching a migration of any kind is to perform a thorough audit of all IT systems to map out application and hardware dependencies. This process is critically important because it will provide an accurate view of how the system is ACTUALLY functioning rather than how it SHOULD be functioning. While one might expect older organizations to have accumulated much more technical debt, the age of the IT system itself should not be used as a shorthand for how difficult the application migration will be. It’s entirely possible for a newer startup to have a chaotic, ad hoc system while a much more complex enterprise has a very accurate map of its dependencies. Once the audit is complete, it will be much easier to identify what challenges the IT team will face during the application migration and what specific problems they will need to solve for it to be a success.
Mistake #2: Moving for the Sake of Moving
Organizations can gain very real benefits from migrating out of legacy on-premises environments and into a colocation data center. In addition to shifting their IT costs from mostly CapEx to mainly OpEx, they also gain access to new connectivity options and more reliable infrastructure. However, if the migration is being carried out without any clear idea of why it’s being done, many of those benefits will be difficult to realize. Indecision and uncertainty can quickly lead to cost overruns, delays, and frustration.
Even worse, a poorly planned migration could severely impact an organization’s flexibility in the future. Moving into a local data center that offers potential cost savings might sound appealing, but not if the transition makes it more difficult to expand assets into an edge data center in another market that presents better growth opportunities. The problems could be even greater if that local data center also happens to be a single carrier facility that severely limits future connectivity options.
“If you are moving things for the sake of moving things,” Reimer points out, “you’re probably not focusing on the future of your business, your product, and your customer.”
Solution: Be Future-Focused
Understanding why an application migration is necessary is almost as important as deciding where to ultimately move IT assets. A technology transfer of any kind should be carried out to accomplish very specific goals, not simply for the sake of doing it. To ensure the move is being done for the “right” reasons, organizations need to clearly identify the reasons why the migration will translate into tangible benefits. For IT personnel, that could mean minimizing manual toil and gaining access to advanced monitoring tools that can drive further performance improvements. Other stakeholders might focus on how moving into a colocation environment will facilitate growth or allow the company to deliver better services to their customers. Identifying these very specific objectives will help to inform not just WHY the application migration needs to happen, but also WHERE assets should be moved and WHAT needs to be done when they get there.
Mistake #3: Move Now, Fix Later
When migrating legacy applications into a new environment, there is often a temptation (or pressure) to complete the move as quickly and inexpensively as possible. That means lifting and shifting applications from one environment into another without optimizing or rearchitecting them first. The significant challenge of executing a migration is difficult enough without thinking about how to rethink or redesign the way a deployment functions, but putting off difficult questions about the future gives an opportunity for status quo bias to seep back into an organization’s thinking.
Once everything is up and running in a new environment, there may not be much appetite to put in the significant work necessary to make the most of a new environment. It’s a mistake that Reimer has seen many companies make: “The old ‘lift and shift first and then come back and take a second pass’ when you do all the hard work, it never happens.” The whole point of an application migration is to put IT systems in a position to take advantage of new opportunities. An application migration should be a transformative effort that enables future growth, not a simple relocation of assets.
Solution: Seize the Opportunity for Change
Rather than focusing the migration plan on “the move” itself, IT managers and business leaders need to think about post-migration goals. Just as moving for the sake of moving is a plan for disappointment, failing to take a realistic approach to how to take advantage of the transition could easily undermine the desire to carry out future digital transformation initiatives. That would be unfortunate, because there is no better time to launch ambitious changes than during an application migration. Transitioning into a colocation data center is the ideal moment to rethink longstanding IT assumptions and put new capabilities into place to promote future-focused growth. While rearchitecting legacy applications can be a difficult process, the potential benefits of doing so during the migration are significant. The opportunities for IT managers to make those changes are few and far between, so it’s important to incorporate them into the baseline migration plan itself rather than putting them off to a post-migration cleanup that might never take place.
Make the Most of Your Application Migration with vXchnge