Mainframe Performance Topics Podcast Episode 11 “XI T’ing”

(Originally posted 2017-03-25.)

This has to have been one of the most trouble-prone episodes we’ve ever done, when it comes to pulling it together. All the issues have been audio[1], not the material.

I think you’ll spot some of those but I think the material is very good, so bear with us. (I don’t think it’s fair to say you wouldn’t’ve noticed if I hadn’t pointed it out to you.)

Anyhow, TSMGO. πŸ™‚

I’ve also learnt a fair amount more about Audacity in the process. Perhaps I’ll write about that some time.

I’m particularly pleased with Anna Shugol recording with me. That one we’ve been wanting to do for some time. And we really do expect to do a follow up later on. Stay tuned! We also want to record with her on another topic; I’ll let people who know her guess what that would be about.

I also thought the Topics topic was interesting, crossing over as it did into both Mainframe and Performance.

And the discussion on the z/OSMF Workflow Editor has got me inspired; I just need a sample of the XML[2] to play with. πŸ™‚

Episode 11 “XI-Ting” Show Notes

Here are the show notes for Episode 11 “XI-Ting”. The show is called “XI-Ting” because, well, it is episode #11 and we had one shot at the Roman numeral to use and took it.


In Episode 10 we talked about the Workflow iOS app, and its role in automation. Just this week it was announced Apple has bought the Workflow app and its developers are now Apple employees. The app is now free on the iTunes Store.

One view of this announcement can be read here: Apple Acquires Workflow – from MacStories


A special hello to Australia and Sweden (“HallΓ₯ !”), who we’ve heard from and has listened to the podcast.

Where we’ve been

Martin has not been anywhere (except for Hursley, UK) since our last podcast.

Marna has been at SHARE in March 2017 in San Jose, California. It was an excellent conference with familar faces and was made even better by the ability to talk about the z/OS V2.3 previewed items.


Our “Mainframe” topic was Marna’s experience in trying out the z/OSMF Workflow Editor. The z/OSMF Workflow Editor is a new function in z/OSMF to create your own Workflow, or change an existing Workflow. The Workflow Editor is available on z/OSMF V2.1 in PTF UI43814 and on z/OSMF V2.2 in PTF UI42847.

Marna pointed out some hints about using this new feature. Some discussion points include:

  • “Folders” for information you need to provide: Metadata, Variables, Steps, …

  • You will always have a correctly produced Workflow definition file. Much easier than the old way (Marna used Notepad++ and kept interating).

  • Your first Workflow will probably use a “template” (file like JCL, script, or exec in a step) that you want to drive. You will probably have “variables” (customized values in that template) that you want the Workflow user to specify. Always make sure you associate your variables with the steps! Otherwise they won’t coordinate together.

  • The open source Apache Velocity Engine is used for variable substitution and conditional directives.

  • There is a very small checkbox for resolving variable substitution in the Workflow Editor. Don’t forget to check it if you are using variables!

  • Remember to remove the first “dummy” step when creating a Workflow from scratch. You don’t need it.

There is a self-directed lab to learn about the z/OSMF Workflow Editor available here which you can run on your own system by using the samples in the Appendix.


Martin had a special guest for a conversation on zHyperLinks, Anna Shugol, IBM Mainframe Technical Specialist.

Martin and Anna talked about a Statement of Direction that was released at the beginning of 2017. This topic is very important because it is an innovative new IBM I/O and Storage technology to improve performance for DB2-centric applications. This is designed to provide dramatic improvements in I/O latency, and change the I/O paradigm.

It complements existing technology, such as Hiper Performance FICON and using large DB2 buffer pools. zHyperlinks is intended to provide short distance (150 meters between the CEC and the storage unit) point-to-point improvements, which are expected to support 8GBs (gigabytes per second) with new protocols.

  • Today, there is I/O to the Coupling Facility, and FICON to disk and tape and times associated with those. zHyperlinks is planned to support improvements in the connect and pend times, with Sync I/O wait times being the dominant DB2 components helped with zHyperlinks. A tool to help with the analysis is expected, along with SMF record evaluation, at a later time.

  • The IBM Storage device required for this solution will be a minimum of DS8880, with up to 16 zHyperLinks being able to be connected.

  • The minimum z/OS and DB2 levels will be provided later.

Stay tuned for more on this topic, as further details are released; Remember this is a Statement Of Direction (SOD) rather than a formal announcement at this stage.


Our podcast “Topics” topic has been sub-titled “Some Assembly Required” ; Not the HLASM that mainframers might think that relates to, but actually something in the same vein.

Marna’s 14-yr old son who is interested in hardware has just built his first personal computer. Marna and Martin talk about how that first computer has a lot in common with a mainframe:

  • Workload (“gaming and intense graphics” for the kid) had to be optimized, with availability and performance in mind.

  • Budget was a big consideration. Some compromises had to be had, but there would be no compromise on the Graphics Processing Unit (GPU).

  • Air vs. Liquid cooling? The CPU and GPU need serious cooling (the way he’s going to run it). Liquid cooling was the better choice, but had to be foregone for air cooling (8 fans: 2 CPU and 6 chassis). Granted those fans are pretty good, and Marna wanted them quiet.

Here’s the interesting thing about Marna’s son:

  • He saved up for two years to buy the parts for this computer. Talk about a kid being focused.

  • He did not learn from any mentors, he learned only from YouTube videos. He had never build a computer before. He had a Raspberry Pi, but shunned it as it was “too software”.

Well, the first smoke test passed fine. The thing to understand? The new generation can understand mainframe concepts and likes them, even if they don’t know they are mainframe concepts.

Customer Requirements

Marna and Martin discussed two customer requirements:

Where We’ll Be

Marna will be at IBM Systems Technical University in Orlando, May 22–26, 2017

Martin has a plan to go nowhere but as always, things change with his schedule.

On The Blog

Martin has published three blog posts recently:

Marna has written one half of one, which is not ready to talk about.

Contacting Us

You can reach Marna on Twitter as mwalle and by email.

You can reach Martin on Twitter as martinpacker and by email.

Or you can leave a comment below.

  1. We know what they are and how to fix them.  ↩

  2. I know more than is perhaps good for me about manipulating XML… πŸ™‚  ↩

Published by Martin Packer

I'm a mainframe performance guy and have been for the past 35 years. But I play with lots of other technologies as well.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: