The key to mainframe DevOps success is in quickly identifying and removing major bottlenecks in the application delivery lifecycle. Major challenges include collaboration between mainframe and distributed teams, lack of visibility into the impact of software changes, and limited resource flexibility with scaling out necessary testing initiatives. Now let's take a closer look at some of these key challenges and how IT departments can address them ...
IT Operations remain essential throughout the DevOps approach. Simply because production infrastructure is a complex and delicate stack of technologies, often with a discreet mix of modern architecture and historical applications. It is often impossible for Development to replicate these types of environments. And that is why quite frequently, an application successfully tested in a qualification environment does not work properly in production. However, it is crucial to validate new developments in a representative replica of the production environment.
So there is a temptation to believe DevOps is doomed to constant failure because of lack of continuity between qualification platforms and production systems. Well, this is probably the case if you merely consider DevOps as a simple top-down approach from Development to Operations. Focusing on producing changes at a higher rate only shifts and grows a bottleneck that sits in production.
As a consequence, DevOps must also be seen as a bottom-up process. In other words, if we consider DevOps, we must also take very serious look at "OpsDev" (even if the reverse acronym doesn't flow off the tongue).
For a successful DevOps approach in practice, Development must position itself as a consumer of turnkey infrastructure environments. IT Operations then adopt an OpsDev approach, and provide infrastructure on demand for all steps of continuous integration – from compilation to qualification, through unit testing.
If DevOps is a radical change in how Development works, OpsDev also revolutionizes common IT Operations practices. It comes with more agile "declarative" infrastructures (described and built from source code), an even more sophisticated level of automation and the ability to provide self-service infrastructure for developers.