Dates/Times - - All webinars are on Wednesdays at 11:00am Eastern
Our webinars vary from 30 minutes to one hour. After each webinar, Peter Enrico and Scott Chapman will be available for a post-webinar Q and A.
Understanding and Measuring Warning Track on z/OS ..... February 14th ..... Sign up has ended. Please see Our Presentations page if you would like to download a copy of the presentation slides
Warning Track Interrupts (WTIs) were first introduced a number of years ago on the zEC12. WTIs are used to maximize physical CPU utilization and improve response time by providing a communication method between the PR/SM hypervisor and z/OS. Essentially, PR/SM is warning z/OS that it is about to have a physical processor un-dispatched, and this gives z/OS an opportunity to remove a unit of work from the processor before the actual interrupt occurs. During this webinar, Scott Chapman will further explain Warning Track. Scott will also review the SMF measurements and whether those measurements are useful in understanding and managing your LPARs.
30th Anniversary of Parallel Sysplex: A Retrospective and Lessons Learned ..... February 28th ..... Sign up using this link
This year marks the 30th anniversary of the announcement and availability of Parallel Sysplex, Data Sharing, and WLM. Let's celebrate with a retrospective why these game changes to the mainframe platform and the world of computer performance management. What were the motivating factors and lessons learned? This presentation will be given by Peter Enrico, who was an original member of the WLM algorithms design and development team. The more 'senior' attendees of this session will listen nostalgically. Newer performance professionals will gain insights into the evolution and maturity of the z/OS platform. This is going to be a great 30-year review and will be full of recommendations and insights.
NEW - just added!
Introducing Pivotor Outlier Detection and Analysis ..... March 12th ..... Sign up using this link
Since the beginning of time, performance analysts have wanted a simple and easy way to find unusual data points in their data with the hope of finding and mitigating unexpected changes before they become significant performance problems. Outlier (or anomaly) detection has been explored for decades with various statistical techniques, usually with mixed results: too complicated to set up and manage, too many false positives, too expensive to perform over large numbers of metrics, or too hard to investigate the results. Pivotor’s new Outlier Detection and Analysis capability utilizes new machine learning techniques and visualizations to address these challenges and empower performance analysts with valuable insight. Existing customers will want to attend this webinar featuring speaker Scott Chapman to gain an understanding of the new Outlier reports and how to use them. If you’re not yet a Pivotor customer, attend anyway to see a demo of Pivotor’s latest functionality.
Batch Initiators – WLM Managed or JES Managed? ..... March 20th ..... Sign up using this link
In the world of batch job management, the choice of initiator strategy (WLM or JES) may seem confusing. What are the advantages of WLM initiators over JES initiators – or is there one? Should I/could I use a mixture? In this webinar, Nancy DiFilippo will discuss what you should consider when deciding which workloads are best suited for WLM initiators.
AI on Z: Exploring Common AI Terms on System Z ..... March 27th ..... Sign up using this link
AI seems the be the big focus and push right now by IBM. The IBM z16 has a built-in AI accelerator. It seems we are entering a brand new world, so let’s try to figure all this out together. During this presentation, Peter Enrico will dissect AI on system Z. The primary topic covered will be the most common terms we hear and what they mean. Peter is no expert on AI on Z, but is willing to dip his toe in the water to help start explaining all this. Terminology is a great starting point.
Per CPU Utilizations ..... April 10th ..... Sign up using this link
When CPU utilizations are reported and analyzed, it is most common that the utilizations reported are for the pool of processors configured to a machine or to an LPAR. Rather than reporting the utilization of each processor, usually we just look at the average across all the online processors. Why is this? Is there any value to analyzing the utilizations of each individual processor? How are the measurements for an individual processor affected by HiperDispatch? During this presentation, Scott Chapman will explore reporting and analyzing individual CPU measurements. It will be interesting to see how the measurements for individual processors differ from reporting the utilization of the pool of processors.
AI on Z: Exploring New SMF Measurements ..... April 24th ..... Sign up using this link
The IBM z16 has a built-in AI accelerator. The IBM mainframe community is expecting many real-world uses of this AI accelerator with new exploitation. As always, the exploitation of new functionality always has a cost and will need to be measured. During this presentation, Peter Enrico will explore some of the new AI measurements and measurement methodology on Z. Peter is no expert on AI on Z, but he is willing to dip his toe in the water to start explaining the measurements and how to measure the performance costs of AI.
If you missed a webinar and are interested in the content, click here to go to the "Our Presentations" page under "Tools & Resources" to download a copy of the presentation slides.
Check out our YouTube channel for helpful videos about Mainframe Performance.
There is no charge to attend any of our webinars.