Why do I point this out?
The reason is simple. When we talk about the challenges surrounding mainframe software development, the problem is not the programming languages or lack of programming skills, but rather a lack of tools that interact with other computing platforms such as cloud, Linux, and Windows. The lack of tools means that we are unable to leverage the skillsets that modern software developers already possess. We hear about the shortage of mainframe-centric development skills on what seems like a daily basis, so perhaps it is time to examine a new approach to the problem.
Instead of trying to bring new developers to the mainframe platform, let’s change the paradigm of mainframe development and bring the mainframe to today’s developers!
Current Challenges to Mainframe DevOps
The challenge of extending mainframe DevOps to reach the levels of automated delivery achieved on other platforms, such as cloud, is well documented. Here is an excellent article by Dave Nicollete about The State of Mainframe Continuous Delivery.
The mainframe industry has fallen into a trap – we continue to develop mainframe-centric solutions for DevOps challenges. Vendors create solutions such as continuous integration (CI) or continuous delivery (CD) orchestrators, testing frameworks, IDEs, automated build tools, and more – all of which target the mainframe.
I believe that in most cases, the mainframe industry will be best suited to adopting industry-standard open tools and frameworks rather than trying to build our own mainframe-centric versions of these tools. How can we use industry-leading technologies such as Jenkins automation server, Gradle and Gulp build tools, MochaJS testing framework, and Git source control tools when working with the mainframe platform?
A Solution in Action
In this blog series, we will introduce the Steel Masters: a modern application development team that is using an innovative, multi-faceted approach to overcoming the problems that we described. The Steel Masters have been busy:
- Modernizing the developer experience when working with an application that traverses multiple platforms and uses languages such as Java for the web, as well as COBOL for transaction processing on the mainframe.
- Building an automated continuous integration pipeline for this multi-platform application.
- Building an automated continuous delivery pipeline for the application.
The solution that you will see in this blog series is enabled by CA Brightside, a new and innovative product from CA Technologies. To learn more about CA Brightside, register and attend our technical webcast on Thursday, June 28, 2018 at 10:00 AM ET.
Chatterjee, A. (2017). 4 Reasons Why Your Choice of Programming Language Doesn’t Matter Much. Retrieved from simpleprogrammer: https://simpleprogrammer.com/programming-language-doesnt-matter/