Many of today’s mainframe applications that were developed in outdated technologies are still mission critical and continue to deliver significant benefits to the organizations that use and rely upon them. Nevertheless, the total cost of ownership and technological risk associated with these applications have now escalated to unacceptably high levels. These applications are candidates to be transformed. The Migratonomy process of Anubex is a method that highly reduces risk from software transformation projects. Because Migratonomy does not introduce any business disruption it becomes possible for an application to undergo a complex cross-platform transformation concurrently with regular application maintenance.
Migratonomy builds on the combination of powerful, flexible CodeTurn tool for source-to-source translation, and the DataTurn for data migration. Furthermore, two automated testing tools are used to verify and prove correctness of the performed code transformation and data conversion: TestMatch for on-line applications and message-based application interfaces such as MQ, and DataMatch for batch applications.
The CodeTurn source-to-source transformation tools use fast(est) parser generators, are programmed with highly optimized generic programming languages, and have powerful command line interfaces. This core benefit translates into the reassurance that if and when, during the migration project, changes are made to your applications in production, these changes can easily be integrated into the migrated system in a matter of hours or days, not weeks or years. Armed with such a capability, organizations are assured the smallest "code freeze" periods imaginable and maximum flexibility for the applications that run their business.
Automation also ensures consistency: both in the code transformation (CodeTurn) as in the data conversion (DataTurn). Additionally, advanced tools to compare the generated source code with the previous generation, ensure that previously transformed code or manually tuned code don't create conflicts between consecutive transformation iterations. Consistency to this level of precision leads to the benefit of being able to work in cycles, without the onerous chore of identifying, extracting, or transforming deltas. With such a capability, organizations are assured a predictable, repeatable process that furthermore gets easier each time it is executed.
Naturally, this means that CodeTurn and DataTurn used for your target infrastructure are also customized to fit your individual migration project.
All Anubex migration projects involve systems that have been in production for decades and have seen multiple changes and upgrades in the underlying infrastructure (hardware and/or software) and sometimes even suppliers. This combination results in an environment that is, by now, quite unique. The same goes for the chosen target technology stack, where the combination of infrastructure and product versions creates a one-of-a-kind system.
A key concept within Migratonomy is the migration staging area: an environment dedicated solely to the migration project. In this staging area a coherent version of the application is installed – this version is called a snapshot and such snapshot includes all the programs, data structures and data. The usage of this staging area ensures the artefacts in development, test, and production are left undisturbed while the migration is in progress.
By working with these snapshots, it is possible for the regular maintenance release cycles and the migration release cycles to happen independently of each other. This separation assures maximum efficiency of both the maintenance and migration processes.
The migration always aims to achieve functional equivalence with the latest snapshot installed in the staging area, and not with the live production system. As time progresses and CodeTurn advances, the snapshot migration involves less work, which makes each iteration naturally go faster than the one before it.
What takes place during an iteration? Each iteration involves the consistent ordering of an identical number of steps.
The first step in any iteration is to take a snapshot of the development artefacts currently in production, a copy is made to the migration staging area, followed by a second step to verify the completeness and correctness of this snapshot by the Anubex CodeTurn analysis tools.
The third step is to transform code and data structures, using the CodeTurn tools to the level at which they have been customized at that moment and to convert the application data. During data conversion, DataTurn creates a new database, and populates it with either (anonymized) data from the production environment, or a specific set of test data.
In the next step, the Anubex testing tools TestMatch and DataMatch measure the correctness of the migrated application. A difference may result in adjusting transformation parameters, the tuning or customization of CodeTurn tools, or perhaps the addition of source artefacts that were missing from the snapshot.
This may start a new “transform-test-tune” mini-iteration, until functional equivalence between this iteration's snapshot and the resulting migrated application is proven. At that point, a new major iteration can start.
The combination of speed, consistency, and customization mean that any change to the original programs' functionality can be taken through to the target platform and regression testing started without the functionality ever being manually identified or analyzed.
Get an instant copy of this Fact Sheet.