In my last post, I mentioned the benefits of using coding standards and today we are going to discuss how important it is to have a clear picture of your source code inventory.
It is quite common for mission critical corporate systems to reach millions of lines of code while being maintained and extended for a life cycle that can go well over a decade.
Using source code static analysis tools in such systems presents not only an opportunity to get an overview of the system’s adherence to coding best practices but it also gives developers an opportunity to easily identify security and performance issues that can be addressed with minimal risk.
There are source code patterns that are easy to refactor and may generate considerable performance improvements.
Some examples are looking for unsigned numeric fields in COBOL used in arithmetic operations or looking for string concatenations patterns in Java.
Source code changes required to address this kind of issues are straightforward and on systems with millions of lines, you may easily find hundreds of these occurrences.
Working on these issues can lead to an overall performance increase and to significant CPU cycles usage reduction, which on Mainframe systems may directly relate to your operational costs.