Open-source software is making great strides on the mainframe. Linux on z Systems is one of the largest drivers
of new z Systems workloads now, and we've also seen IBM embracing even bigger ideas such as OpenStack
. However, traditional z/OS workloads have largely not been impacted by open-source products or ideas. I think this is about to change, and it will change faster than it ever has before.
Open-source software is written by anyone and made available for sharing and collaboration under a variety of software licenses. The licenses range from being virtually unrestricted, limited to non-commercial use or have other restrictions. The variety of open-source software that is currently available is massive and growing every day.
UNIX System Services (USS), a component of z/OS, has been around for 20 years (introduced as OpenEdition MVS). Its presence has worked its way into a central role for z/OS and many z/OS applications. However, USS has not fully blossomed. For starters, the open-source community and the z/OS programming community don't overlap as much as they could, and the tools available for USS has been historically limited and few in number.
Working in a z/OS environment requires some specific skills. From endian or codepage issues, z/OS doesn’t share many of the same building blocks as more commonplace x86 architectures. It requires some work to make open-source software run under z/OS, and it takes more work to let the software reach your z/OS data. This can be interpreted as a good thing because organizational core business data lives in z/OS, and the standards for working with it are set high.
The z/OS platform needs understanding and advocacy that spills outside of its dedicated user base. Any of these open source packages can run on z/OS with the right changes and testing. A great example of this is Perl on z/OS. With some corporate assistance, a small group of z/OS developers who love Perl and z/OS have helped create and maintain Perl in the official source tree for Perl worldwide. Going forward, advocacy, testing and modest support will make sure future z/OS programmers have access to Perl today and in the future.
This is the model that must to be replicated. We are z/OS, and z/OS needs to be a player in the open-source community. The open source community is just that: a community. It is a large community that is interested in sharing with the z/OS community. Unfortunately, it is a community where z/OS has not been actively involved.
There are millions of pieces of useful open-source software available today that could drive z/OS to new heights. Minor awareness of z/OS could enable those, and many more, to run on z/OS.
So here is the call to action: If a piece of software doesn't run under z/OS in some form, step up to make it run on z/OS. Talk to the maintainers; port it yourself; ask someone to port it for you. Above all else, represent z/OS to the open-source world. There are too many cool things we could be doing with z/OS with a little help from our friends.
Kevin Shaw, a UNIX/Linux nerd at heart, manages software, development and partner relationships for the z/OS Storage software business at Rocket Software. He has experience leading development teams in z/OS database and z/OS security products. He was once a mainframe developer at a large financial institution.