March 14, 2017 | Written by: Chris Sayles
Share this post:
You know the saying, “an ounce of prevention is worth a pound of cure”? It’s a famous after-the-fact phrase, meaning it is a phrase you often hear used by folks outside the situation, looking in and saying “I told you so.” But nonetheless, no matter who is using it and when, it is often very true. When we apply this to modern development practices with regard to shift left testing, it becomes even more poignant. The reasoning for this shift (pun intended) is that we want to create a culture of prevention rather than detection. We want to catch the bugs early, chase them down quickly and resolve them efficiently. What we don’t want is to have to damage-control our apps when they crash, causing downtime and high resource utilization, both of which are costly and detrimental to your business.
A focus on quality
To ensure quality code, there needs to be tooling to support a more thorough code review and governance process. IBM’s Developer for z Systems offers a number of code quality features like content assist, code rules for code review, and integration with a robust, source-level debugger. Developers are able to deliver quality code at a fraction of the cost. Add in Application Discovery and Delivery Intelligence to bridge any gaps in code base knowledge, and it’s much easier to identify where components can use updating, modifying, or even new development.
Through data collected, correlated, and calculated, IDz and ADDI produce invaluable application and testing health metrics that allow teams to see trends in testing results. ADDI’s Delivery Intelligence component provides coverage trend analytics at the application and program level. It can help you understand whether you are improving your test coverage over time, gives you immediate insight as to how much of newly added code or changed code is being tested and/or covered and can recommend sets of tests to run for a given set of changes in a build by leveraging the code coverage data from previous test runs.
Testing, not detection
If you don’t want to damage control your way to a successful deployment, testing frequently and accurately (read: scoping properly) is a must. IDz offers an automated unit testing facility called zUnit that allows for the creation of test cases for COBOL and PL/I batch applications, giving developers a way to really hone in on individual program source files and run them against a particular set of variables to understand how well the source is performing. This can be done as often as needed to ensure accurate results with new builds of your applications.
Shift left testing doesn’t have to be a massive undertaking. With tools like IDz and ADDI, making the shift to a development practice that saves your shop time, money and frustration is very easy.
Click here to watch videos of IDz and ADDI in action.