Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Five steps for successful app modernization and legacy migration


Legacy application migration checklist

Source:  Sergey Nvns/Fotolia

It's not uncommon today to find 20- to 30-year-old legacy systems that have been upgraded and re-patched. Banks, government agencies and other businesses rely on these systems to perform daily customer and business transactions. Some systems aren't able to keep up with ever-increasing demands, particularly when it comes to large amounts of transaction and data. As a result, many software architects' and CIOs' agendas are weighing the risks of legacy application modernization.

To keep up with technology changes, some slow-performing and resource-intensive old systems have been updated, and others are being considered for legacy application modernization on new platforms. The latest platforms deliver resource optimization, higher scalability, faster data speed and quicker workload balancing over the Internet.

One legacy migration approach is to repurpose old applications into resource-efficient and optimal applications in-house or for the cloud. If a legacy application has been successfully modernized in-house, it doesn't mean it can also be used as a Software as a Service, or SaaS, application.

Not all applications that run successfully in-house can be migrated to the cloud. In order for the migration to be successful, the behavior of the application needs to be changed for use over the Internet.

Changing application behavior come with risks, however. The dangers include slower performance; generation of garbage output; poor scalability; high resource consumption; frequent application failures; and in the worst possible scenario, system crashes.

Follow this application migration checklist to mitigate risks during your legacy modernization effort.

View All Photo Stories

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

Article a little lean on detail. If an app is running 20 or 30 years after it was written, that is a tremendous ROI. As for faster data speeds, the "legacy" platforms tend to have the highest data speed and far better multiProcessing (ie: processor becomes free to do other work while IO is occurring. There are times to migrate, but I don't know that much of a case was made here.
This approach doesn’t sit so well in context with SearchSOA. I would suggest different steps:
1. Assess rational for modernization of the cluster of applications that support chosen business capabilities
2. Assess application capabilities relationship to service portfolio, including solving duplication issues and standardization opportunities
3. Define architecture that facilitates transition and retirement as necessary
4. Plan “baseline” modernization that delivers services and APIs as per portfolio
5. Execute knowledge discovery to automate transformation where possible
Continuously evolve service functionality to support changing business needs
Thanks, DSprott. You've definitely shared some good suggestions!
Legacy systems can be the hardest to migrate over, especially with systems that are so reliable. May need a fresh approach, when dealing with systems that are coded for so long to function in a certain syntax.
I would like to know if there are any open source examples of migrated legacy systems/apps to SOA
Hi Ellyne,
We are migrating a core banking system to micro services and JBoss Fuse (open source) and have built a related methodology, contact me for details