Process for Mainframe Migration

The Value of a Good Discovery Process for Mainframe Migration

  • Twitter
  • Linkedin

Much of the focus of IT modernization presumes an organization has an existing install base of computing infrastructure and applications and wants to move them forward in some meaningful way. Modernization that starts from a clean IT slate is a different animal all together. I remember an architect friend of mine (building architect not an IT one!) telling me he preferred designing new buildings over re-architecting an existing one! Starting from a clean slate means no pre-existing conditions/restrictions. But most legacy or IT modernization discussions I’ve seen over the years presume that the starting point IS existing applications and the infrastructure they are built on.

Organizations are faced with two options. Take the “build it new” option and incur the large costs and risks and hope for the best. Or take a more cautious, albeit seemingly much less gratifying path of incremental modernization. You don’t get to your desired end state directly, but it does give you increasing value over the course of your modernization journey. And that’s what the LzLab’s customer journey approach is all about. When I was in university, I had a poster of two vultures sitting on a cactus in the desert. One was saying to the other “Patience my ear. I’m going to kill something!”. Well it went a little different than that, but you get my point. This is the kind of attitude IT modernization projects often exhibit. A lack of patience inclines an organization to go straight for the kill – write a new application from scratch or migrate the business process to a packaged software solution. There are certainly some cases where this approach is warranted, but when taken as an organizational strategy for all IT modernization projects, technological mayhem may ensue!

Picking the right road

My father-in-law had another famous saying, “If you don’t know where you are going, any road will do.” But IT modernization projects should know their destination and picking the right road is important. But before you start down that road, you need to know from where you are starting. The first step of any engagement with LzLabs involves a discovery and assessment phase. Many organizations struggle with this step for two reasons. First, it doesn’t seem like a good use of time. Secondly, many are unwilling to admit they really don’t know everything they need to know about what the current application portfolio is all about. When new CIOs enter the picture at an organization it’s much easier for them to ignore these two reasons – just blame it on the last CIO!

Once an organization has a good understanding about the structure of their application portfolio and the interfaces that have been defined to support the evolution of the changing business process over the years, they can better define the iterative steps necessary to modernize to their desired end state. The desire for more immediate gratification drives many application development teams to want to skip this step. After all, they know everything they need to know about the application they support! What they know is the 20% they’ve probably supported the most in the last decade, but much less about the components that have just run for years.

Understanding the underlying technologies

The LzLabs LzDiscover step is designed to highlight the application components affected by a modernization project. This includes the underlying technologies that were used to implement the application over several decades, as well as many of the specific features that were used. An understanding of the underlying data model and how it’s shared across applications is critical to defining a successful modernization migration plan. LzLabs and/or our implementation partners use this information to lay out a transformation plan that ensures that now only all the correct parts are migrated, but also that any technical or business interfaces are preserved. Lastly, this effort can help organizations understand how they can distribute the application in a more meaningful way across a distributed environment. There is no need to preserve the “scale-up” mindset of the mainframe. Once the parts of an application that can be split are identified, the implementation on the LzLabs Software Defined Mainframe positions the organization for improved testing and quicker and lower risk innovation steps.

In subsequent blogs, we will focus on the benefits gained from the next steps in the customer journey, when migrating mainframe applications with the LzLabs Software Defined Mainframe.