Design and implementation of quality assurance measures for host online and batch systems. Technical changes to the mainframe system environment result in different application characteristics, e.g. poorer performance, program terminations, different calculation results. The changes are intended to improve the performance and remove errors from the software system operation.
Supplement
Difficulties arising from patches and other maintenance work on the system environment occur more frequently in older PL1 modules (some of which are more than 20 years old), which have rarely been altered since they were first introduced. The various different error types require different solution methods. An iterative and multi-stage procedure is useful for problems that occur in a large number of programs and that must not disrupt normal operation when they are introduced after revision. The instruments used for quality assurance include utility programs on the mainframe (e.g. ICETOOL, IDCAMS, DFSORT) and tools on the PC (e.g. Perl scripts, HTML, Office products).
Subject description
The issue of quality assurance for mainframe online and batch systems involves a series of different technically motivated activities of varying complexity that are all aimed at ensuring operational reliability and improving performance. The problems are rectified either automatically or manually. Automatic correction is useful for difficulties that occur in many programs with a certain frequency, e.g. software features that are no longer supported. Perl scripts play a significant role in the automated analysis and problem elimination. By contrast, manual fault correction can be used for problems that occur in few modules and that require more complex program adjustments or even new implementation. In batch processing, quality assurance also involves job streams in addition to the P?L1 programs.