Python Powers Modern Mainframe
Key Points
- Python’s extensive data‑science ecosystem runs natively on the mainframe, giving scientists direct, high‑speed access to the 70 % of the world’s structured data that resides there and enabling inline model execution on the Telum processor.
- Site Reliability Engineers can leverage the same Python tooling they already use for infrastructure‑as‑code to automate and manage z/OS environments, calling legacy REXX/JCL when needed while exploiting mainframe hardware features such as built‑in compression.
- CIOs benefit financially because mainframe workloads can be off‑loaded to the zIIP processor, which incurs no additional software licensing fees, delivering extra processing power without increasing the organization’s cost base.
- In short, the familiarity and power of Python seamlessly extend to IBM Z, turning existing skills into productive mainframe capabilities across data science, operations, and cost‑management domains.
Sections
- Mainframe Meets Python Data Science - The speaker argues that Python‑savvy data scientists can leverage mainframe‑hosted data and the Telum processor to run models inline, combining familiar tools with massive, fast‑processing workloads.
- Cost‑Free Mainframe Python with zIIP - The speaker explains how CIOs can adopt Python on mainframes cost‑effectively using the licence‑free zIIP processor, which adds capacity without extra charges.
Full Transcript
# Python Powers Modern Mainframe **Source:** [https://www.youtube.com/watch?v=s75Ysq38TZE](https://www.youtube.com/watch?v=s75Ysq38TZE) **Duration:** 00:04:04 ## Summary - Python’s extensive data‑science ecosystem runs natively on the mainframe, giving scientists direct, high‑speed access to the 70 % of the world’s structured data that resides there and enabling inline model execution on the Telum processor. - Site Reliability Engineers can leverage the same Python tooling they already use for infrastructure‑as‑code to automate and manage z/OS environments, calling legacy REXX/JCL when needed while exploiting mainframe hardware features such as built‑in compression. - CIOs benefit financially because mainframe workloads can be off‑loaded to the zIIP processor, which incurs no additional software licensing fees, delivering extra processing power without increasing the organization’s cost base. - In short, the familiarity and power of Python seamlessly extend to IBM Z, turning existing skills into productive mainframe capabilities across data science, operations, and cost‑management domains. ## Sections - [00:00:00](https://www.youtube.com/watch?v=s75Ysq38TZE&t=0s) **Mainframe Meets Python Data Science** - The speaker argues that Python‑savvy data scientists can leverage mainframe‑hosted data and the Telum processor to run models inline, combining familiar tools with massive, fast‑processing workloads. - [00:03:03](https://www.youtube.com/watch?v=s75Ysq38TZE&t=183s) **Cost‑Free Mainframe Python with zIIP** - The speaker explains how CIOs can adopt Python on mainframes cost‑effectively using the licence‑free zIIP processor, which adds capacity without extra charges. ## Full Transcript
You love Python.
I know you love Python.
Now, I love the mainframe,
and the mainframe loves Python as well.
So give me just a few minutes and I'm going to explain why
your love of the Python will translate to your love of the mainframe as well.
Here we go.
Let's say you're a data scientist.
As a data scientist, you care about more data and fast processing.
Well, let's think about it.
The mainframe today has 70% of the world's structured data, at least, on the system.
So you have access to large amounts of very important data
that runs banking systems, insurance systems, the financial world.
Using Python, the Python you're used to,
current versions of Python, the most popular Python packages that you use for data science
are available on the mainframe as well.
So you can use your skill, your knowledge - and you can apply it to all of this data.
And for fast, we have the Telum processor.
The Telum processor allows you to load those models into the system
and allows them to run alongside the existing processing.
So your job: make sure fraud detection happens.
Well, I can do that inline as part of the processing.
I don't have to wait, I don't have to call out to some other system.
It's inline processing and I can do it as fast as a credit card swipe.
So as a data scientist, I use all of my existing tools,
all of my existing capability, and I get access and faster.
Now let's turn to the next person, the SRE.
When I think about SREs, you're using Python today, you're doing infrastructure as code with Python.
It's the most common language when we're dealing with SRE environments
so that I can manage systems effectively.
Well, on the mainframe, you may think about REXX and JCL,
but forget about that.
You can use Python to do all of what you need to do on z/OS.
You can ignore those languages and just build new things using Python.
And if you have to, you can call existing REXX and JCL if you want to.
Or you can do it all in Python.
Familiar.
It has full capability to work with all of the existing z/OS systems
so that you can configure your middleware, do all the automation necessary.
And you get to take advantage of the existing hardware
so you can use things like the compression available on the system
through Python.
So you're not having to do something different or call something different.
You have access to the hardware itself,
capabilities from your native Python.
And now when we think about this last use case, the CIO,
what do CIOs care about?
Money. What is it going to cost me to do this?
Well, in the Z world, we have the zIIP processor.
The zIIP processor is running alongside the general purpose processor.
And with the zIIP, you don't pay a monthly license charge.
So you're getting to do all this additional processing
without adding to your bill and your bottom line.
So this makes all of this work efficient, easy, effective,
and without costing anymore.
So hopefully it's clear
that you love Python and now
you can love the mainframe with Python
since the mainframe is what runs the world.
Thanks for watching.
If you like this video and would like to see more videos like this, just click like and subscribe.