Offer
The modernization of an existing IT system means usually a huge change. This can be done in several small steps or one big step. Platforms, runtime environments, and programming languages are often changed in large migration projects.
Analysis
In the analysis phase, we get a certain overview. We assert ourselves in detail with the goals of the modernization project, the temporal, organizational, budgetary boundary conditions and targeted analyses.
TECHNICAL ANALYSIS
We take a closer look at the legacy system with the help of static code analysis. The result are e.g. responsibility graphs, size information, recognize recurring code patterns and the identification of unfamiliar codes. This information is the basis for improving a practical, goal-oriented modernization strategy.
USER-CENTERED ANALYSIS
With the help of context interviews, use cases, contacts and other usability engineering methodologies, we investigate more in detail the user's perspective of an existing legacy system. This means for us more information for prioritizing and planning the modernization project step by step. One of the most important factors for a successful migration is a full user acceptance of the new system.
Concept
There are many reasons for modernizing existing IT systems. Nevertheless, such projects are often pushed into the background. Hurdles are often high investment costs and various organizational challenges.
Often, basic processes have to be redesigned, and an organizational change has to be achieved at the same time. The workforce must accompany this change and learn more about methods, processes and technologies. In addition, the old system not only has to continue to operate during the modernization project, but also has to be in parallel maintained. Some basic success factors for modernization projects provide orientation for overcoming individual challenges.
For a major customer in the financial sector, we migrated batch processing with around 1,800 interfaces away from hard-to-maintain mainframe code to a modern architecture. The old system was developed with PL / I and JCL.
Serano Colameo
CEO, arcware GmbH
Preparation for a Successful Migration
Changes require good preparation! The removal of unused code parts reduces modernization and maintenance costs. Increased test coverage and automation of tests ensure shorter development cycles and prevent regression errors. Improved modularization of the old software enables step-by-step modernization and ensures optimized testability
Proof of Concept
Modernization projects can be technically and organizationally complex and challenging. In a proof-of-concept, feasibility is checked prototypically at manageable costs, alternatives are analyzed and expected expenses are compared with the benefits to be achieved. This also determines which parts can be automated at a reasonable cost. The results are concrete recommendations for action and a project plan for the project.
Tool Support
Good tools help with many tasks in modernization projects. Analysis tools ensure sufficient detailed knowledge of the legacy system. Tools for test specification and automation, improve the initial situation. Automatic transformations of code, tests or data from the old system ensure your investment. Code generators increase quality and efficiency and reliably implement architectural specifications. An important success factor is the adaptability of these tools.
Team Work
A good team is an important basis for every successful project. Different skills are required when modernizing old systems: a good understanding of the technologies of the old system, detailed experience of the target architecture and the necessary soft skills for change management. We support you with individual development tasks or take on complete projects.
Model-based Migration
The old system is a valuable investment. As much as possible of the existing old system needs to be extracted and used to create the new system. Therefore, tool-supported analysis of the legacy system is carried out at the beginning with the aim of recognizing structures in the code and in the data management and of extracting them as much as possible automatically.
Outstanding functions of the analyzed legacy system are developed into formal models with complex abstraction. The models cover various aspects (front end, business logic, data, etc.) that are automatically extracted from the legacy system. Further, models for checking the system description can be managed manually. The models are then used as input for the generator of the target system.
The (rough) target architecture must be determined before the generator is adapted. All aspects of the target architecture are implemented manually in the form of a reference implementation. The assignment rules (templates) for the generator are derived from the reference implementation.
The generator produces code and artifacts of the target system using the templates. However, the generated target system is enhanced and adapted where needed manually. The successive refinement of the target system is carried out in an iterative process for a specific refinement of the generator.
