Friday, August 21, 2015

Python on Big Iron: A Challenge for the Future

Python is a fantastic language. It can do just about anything you need, it allows you to write code in a manner that's comfortable for you, and it stays out of your way as much as possible. Over the last few years, I've found myself turning to Python more and more instead of 'traditional' languages like C++, C#, or Java. Python allows me to be more productive than most of these languages, is cross-platforn, and, on modern hardware, it's usually 'fast enough'.

One place where Python falls down hard, however, is on so-called 'Big Iron'. I'm talking about those massively powerful computers used in various industries like the financial, medical, entertainment, and others that require solid, reliable, scalable performance. Mainframes.

While we usually don't think about it, mainframes are still heavily in use around the world. Chances are pretty good that you interact with these beasts every single day and someone is writing code for every single application that runs on them. Unfortunately, that code is not being written in Python. It's written in a variety of other languages like Java, C++, and COBOL, but it's not, almost ever, Python.

Python support on modern mainframes is abysmally shameful. The last time I looked at doing a mainframe project, the latest version of Python I could find was 2.4.1. That's old. Like 2006 old. There's a lot of talk about bringing 2.7 and even the 3.x branches of the language to the IBM mainframe, nobody seems to be making any moves to actually do that (and I'm looking squarely at you, IBM).

Don't get me wrong, I understand that mainframe development is pretty niche. But it's still another mountain that I believe Python should climb and could shine on. It's time our community understands that there's a lot of code being written for the mainframe and we're losing that battle. The mainframe isn't going away anytime soon. In fact, IBM has seen an uptake in sales of its z-Series and OS/390 systems and, as data becomes more and more complex, I suspect the need for large systems is going to increase.

With all the complexity the mainframe brings, opening development to potentially millions of developers by supporting a modern version of Python would benefit both the language itelf and the mainframe vendors themselves. It's an awesome opportunity that I truly hope that both communities focus on in the very near future.

1 comment:

PatriciaMoncrief said...

It looked fascinating and I chose to investigate each substance and I read yours. My compliment about this blog is exceptionally positive I visit this blog first time and inspire by this good stuff work. Incredible post keeps up posting such great information. If you like to travel then you can use munnar call taxi for a trust worthy travel.