Recompiling complex applications is often manual, time consuming and error prone
Recompiling applications and correctly taking into account the impact on dependant components is often a manual, time consuming and error prone process. Traditional big-bang approaches involved recompiling everything and therefore did not need such an intimate understanding of component interdependencies. Today however there is a business demand for more frequent application deliveries in smaller increments. Understanding the interdependencies of sub sets of impacted components has never been more important. Correctly applying pre-requisite actions and correct parameters is a weakness of the manual process but it’s critical to get this right to maintain quality, availability and compliance.
Providing frequent application improvements requires first class build automation
It can be difficult to correctly recompile individual components. There may be specific actions which must be performed before and after recompile. There may be specific, non-default recompile parameters. ILE, SQL, multi-system and multi-platform environments introduce more dependencies and have made this even more complex.
Improve quality and save time when recompiling application components, particularly in modern rapid delivery environments
Build automation must:
Automatically reference a self-maintaining repository which holds current information about all application components.
Ensure that only dependent components are included in the build.
Ensure that all essential pre and post compile steps are introduced into the build.
Ensure that all the correct parameters are used in each recompile instruction.
Ensure that the recompile sequence is correct.
Ensure that authority and ownership is preserved.
Ensure that data is transferred and correctly mapped to new data structures.
Do this all automatically and get it right every time.