Started as a weekly series of 15 webinars during the beginning of the COVID pandemic, it is now a household name in the mainframe world. The IntelliMagic zAcademy is a monthly conference-quality educational session for everyone within the z/OS community.
Date: February 23
Time: 2 PM ET | 1 PM CT
One of the greatest challenges in IT today is balancing speeds across various components in your systems. Storage of all flavors has grown bigger, but not faster, compared to CPU speeds.
Your systems read (and write) enormous volumes of data – perhaps terabytes per second. They access data from multiple levels of processor caches, main memory, system paging areas, coupling facilities, disk (now largely solid state), and tape (often disk!). These areas vary dramatically in size – but more importantly – they are not accessed at the same speed!
Today’s CPU’s process at over 5 billion cycles per second and we all know the adage: “all CPU’s wait at the same speed.” The key is to make as many of those cycles as productive as possible.
This session will discuss the many tiers of storage in IT systems and offer ideas about how to optimize access to those areas.
Date: March 23
Time: 2 PM ET | 1 PM CT
The mainframe runs mission-critical work. Performance and capacity teams work tirelessly to avoid preventable service disruptions and to minimize expense, seeking to achieve this despite decreasing staffing levels.
There are few mainframe teams who don’t have experienced personnel closing in on (or already surpassing) retirement age. These tenured staff members are often the last line of defense supporting legacy home-grown reporting tooling. Yet staff shortages and demanding workloads combine to minimize the success rate of skills transfer initiatives, however well-intentioned.
Without the ability to retain experts to maintain in-house tooling and to interpret the increasingly complex metrics, the likelihood of avoidable service disruptions and their related business impacts increases significantly.
Business as usual approaches are unlikely to succeed in the face of today’s widespread challenges of “brain drain” from retirements and shrinking staffing levels. Attendees of this webinar can expect to learn proven strategies being leveraged by companies to optimize mainframe availability and performance.
Presenter: Todd Havekost
The mainframe produces a set of metrics far superior to any other platform. Unfortunately, there are significant barriers to unleashing the learning value of this SMF data.
Attend this session to learn how rapidly you could leverage easy visibility into RMF and SMF data to expand your understanding of how the z/OS ecosystem operates, and renew your passion for the disciplines of performance analysis or infrastructure support!
Request link to watch recording
Presenter: John Baker
The performance analyst is that rare breed that simply must know – in excruciating detail – what makes things take so long, whether the “thing” is a DDF transaction from the web or a line at the store.
Regardless of your job, we all hate to wait, so understanding these principles can be quite satisfying. Take a stroll with John as we shed some light on these concepts. No degrees, certifications, or programming necessary – this intro session is for everyone.
Request link to watch recording
Presenter: Jack Opgenorth
You’ve already heard some good advice on optimizing HSM, but what about the new z15 features? What about your different tape compression options? Or changing your storage management strategies?
We’ll use some comparative data to outline some opportunities to consider going forward. Previously referenced recommendations will be emphasized, and other good practices will be highlighted to help you rein in your HSM demand.
Request link to watch recording
Presenter: Todd Havekost
This session will present the benefits of modernizing how you understand and leverage SMF data for CICS and Db2 Performance analysis.
Instead of being overwhelmed by massive volumes of CICS and Db2 SMF data and combing through countless static reports, discover how interactive analysis using an intuitive GUI interface with context-sensitive drilldowns can elevate your effectiveness to an entirely new level.
Request link to watch recording
Presenter: Todd Havekost
Since it is likely that 1/3 to more than half of your total CPU cycles are spent waiting for data to be staged into level 1 processor cache, it is critical to understand and have visibility into this significant component of your CPU consumption.
This is not ivory tower theory… every main point in this presentation is backed up with real-life case studies from more than 70 sites across 8 countries.
And though these concepts have been presented before, the presentation continues to be enhanced with additional before-and-after data from sites quantifying the CPU reductions they have achieved through more effective utilization of processor cache.
Request link to watch recording
Presenter: John Baker
In IT, the term “Flash” is both ubiquitous and vague. The solid-state drive in your laptop, the new memory in your z15, your new all flash storage array – these all boast flash technology that promises to make everything faster.
Many datacenters have adopted flash technology and are observing that their I/O response times are not dramatically different. Why? John breaks down this relatively old invention to understand what it can – and cannot – do for your datacenter.
Request link to watch recording
Presenter: Todd Havekost
Given the importance of security in today’s world, managing network encryption is a critical requirement for almost every z/OS environment.
I consider the introduction of zERT SMF 119.12 records to be perhaps the greatest advance in SMF data since CPU Measurement Facility (CPUMF) began providing instrumentation of processor cache metrics through SMF 113 records in 2008.
Attend this session to learn more about the value provided by this new zERT data, and to gain the visibility your security team needs to derive great value from this data.
Request link to watch recording
Presenter: John Baker
There are few things more annoying than a daily job running dramatically longer for no apparent reason. Such delays can trigger financial penalties leading to pressure on the performance team.
Performance Analysis is about identifying the distribution of response times and dealing with each component. In this presentation, we will explain the meaning of some the most useful components and provide practical recommendations on how to improve application and overall system performance.
Request link to watch recording
Presenter: Todd Havekost
IBM’s MQ is very widely used across today’s z/OS environments. But reporting capabilities for MQ SMF data have lagged behind those of other subsystems.
This session will present the benefits of modernizing how you understand and leverage SMF data for MQ performance analysis.
Instead of being overwhelmed by massive volumes of MQ Accounting data with very limited visibility, discover how you can elevate your effectiveness to an entirely new level.
Request link to watch recording
Presenter: Dennis Moore
WebSphere Application Server is mission-critical for many businesses today. However, for many mainframe performance/capacity specialists, these workloads can be a bit of a mystery – a black box of resource consumption that is hard to get your arms around.
There are many areas of WebSphere that can leave even a seasoned z/OS performance specialist scratching their head. Knowing what you want to see and where to look is one hurdle to overcome, but even then, how do you access that data? And more importantly, how do you visualize it in a way that makes sense? This webinar will help you:
Join us to learn how you can shine a light into this otherwise black box!
Request link to watch recording
Presenter: John Baker
The z14 and z15 support a completely new I/O method called zHyperLink. This is easily the most substantial I/O announcement from IBM since the introduction of FICON. We all know that the best I/O is no I/O but every record won’t fit in your buffer pools, so I/O is a fact of life. What about more cache or flash drives? The many options to accelerate your speed on the I/O highway can make you lose your mind – come check out the fast lane!
Request link to watch recording
Presenter: Cheryl Watson of Watson & Walker
Every IBM software pricing option has a set of customers that it will benefit. For the remaining customers, one of the other options might make more sense. IBM’s Tailored Fit Pricing option had some aspects that limited the set of customers that it was a good fit for. In this webinar, mainframe maven Cheryl Watson describes recent, unannounced, changes to TFP, and explains how those changes might alter your calculation about whether TFP is right for your company.
Request link to watch recording
Presenter: Frank Kyne of Watson & Walker and Todd Havekost of IntelliMagic
Most installations tend to upgrade their CPCs to the equivalent model in the newer generation – for example, from a z13 710 to a z15 710. Accordingly, IBM’s capacity planning tools are designed to model upgrades to newer generations within the same speed range (7xx to 7xx, 5xx to 5xx, and so on). There are numerous stories of customers moving to a lower speed range (z13 7xx to z15 6xx, for example) and finding that the performance exceeds their expectations. But if the IBM tools are not intended to model such moves, how do you evaluate if such an upgrade would be right for your workloads? In this session, Frank and Todd will discuss the attributes that they use to evaluate such upgrades.
Request link to watch recording
Presenter: Todd Havekost
With all the complexity and workload variability present in today’s z/OS systems, it is important to have significant deviations of key metrics from their historical norms called to your attention. Built-in “change detection” functionality to automatically identify these deviations accompanied by intuitive capabilities to investigate the underlying causes provides another important step forward facilitating effective availability and performance management. This session will show how this capability complements health assessments using built-in instrumentation of z/OS best practices and will show real-life examples of how it can add value across several areas of the z/OS infrastructure, including CPU, CICS, Db2, CF, Disk replication, and batch jobs.
Request link to watch recording
Presenter: Katja Denefleh
You have multiple functions and parameters at your disposal to define which logical volumes will remain in the TS7700 cache, regardless if you use a TS7700 Disk-Only or a TS7700 with Tape attach. This session gives you detailed information about the different options. Learn more about Cache Partitions, Retention Policies, Preference Groups, Autoremoval, and parameters like COPYFSC and RECLPG0.
We will discuss how these parameters control the cache content and how they may influence recovery times. After this webinar you will be able to optimize the cache usage.
Request link to watch recording
Presenters: Cheryl Watson of Watson & Walker and John Baker of IntelliMagic
Tailored Fit Pricing is no longer an abstract concept as many data centers have adopted this new pricing model. For those that held back, recent enhancements from IBM have made TFP potentially more attractive. While TFP has the potential to greatly simplify software costs the Capacity Planner/Performance Analysts (CP/PA) needs to adjust many of their tried and true methods for tracking – and controlling – MSU consumption. Are you paying too much? Have your measuring tools and techniques adapted to these new rules? Join Cheryl Watson and John Baker as they dive into this critical subject. They’ve been known to have an opinion or two.
Request link to watch recording
Presenters: Brent Phillips and Jerry Street
2020 will undoubtedly go down as a year to remember for each and every one of us. The way that every business has adapted to the changes shows resiliency and creative professionalism in positive ways. One thing that has not changed is the need to ensure that our infrastructure and applications can handle peak loads at any time. This webinar will demonstrate how to quantify risk in the z/OS infrastructure to ensure better availability in peak periods, and how intelligent navigation through the RMF/SMF data can accelerate problem resolution times and display newly supported SMF data.
Request link to watch recording
Presenters: Dan Speer and Todd Havekost
Analyzing and optimizing performance and cost across the different z/OS infrastructure areas has led to many “silos” of expert knowledge which often create natural barriers to collaborative and quick problem-solving. It also makes it difficult for managers and LOB executives to easily understand the performance health and cost optimization status for their applications. This webinar will discuss the historical reasons why these silo barriers exist, their ramifications, how to achieve greater transparency and collaboration, and a roadmap to help break down the silos one application at a time.
Request link to watch recording
Presenter: John Baker
Your mainframe has the capacity to service many requests concurrently. However, as with any system, there are limits. As utilization increases, queues will inevitably develop. This session will look at the applications that are impacted, and whether you need more hardware or can optimize with tuning.
You don’t need to be a mathematician or programmer to answer these “queueing theory” questions. Join this session and learn how IntelliMagic Vision can provide clear answers to these and other questions in your z/OS environment. With direct visibility into delays and the reasons for them, as well as predictive analysis, who needs theory?
Request link to watch recording
Presenters: John Baker, Todd Havekost and Brent Phillips
Last year, over the course of 19 zAcademy educational sessions, we learned of some of the most pressing issues affecting data centers, ranging from CPU Consumption and new pricing models to performance issues in Db2, CICS, and systems components, to configuration best practices for an optimal environment.
Learn best practices for configuring your environment for optimal performance, selecting the best pricing model for your site, and how to navigate performance issues in your Db2, Systems, CICS, and other z/OS components simply by accessing the RMF and SMF data available to you.
Request link to watch recording
Presenters: John Baker, Todd Havekost and Brent Phillips
Our panel of experts continues the discussion from February 9th, addressing 2020’s biggest challenges to infrastructure operations. They draw on real-world experiences at many mainframe sites around the world, and where appropriate illustrate their responses with demonstrations using intelligent and automated analysis of RMF and SMF
Topics to be covered in Part 2 of this session will be based on input from registrants, along with overflow topics from the first session, including:
Request link to watch recording
Presenter: Katja Denefleh
Understanding your TS7700 environment is crucial to ensuring you get the most optimal performance out of it. As a follow-up to last year’s zAcademy virtual session on Cache Capacity, this Virtual Tape webinar focuses on Cache Bandwidth.y
Learn more about cache flows, delay premigration, the new premigration behavior introduced with TS7700 Microcode 8.5.1.25 (available for TS7760 and TS7770), and the influence of some features like compression or copy policies to your cache bandwidth consumption. To provide the full picture, library backend performance and physical tape drive usage are also discussed.
Request link to watch recording
Presenters: Jerry Street and Jack Opgenorth
Contrary to the misguided public view, modern mainframes are nowhere near antiques, boasting the best performance, encryption, compression, reliability, and instrumentation modern processors have ever experienced.
This webinar will cover in detail the modern-day equivalents of yester-year’s antique performance tools. We’ll also cover:
Don’t be surprised if we show you something in your toolbox that can be improved!
Request link to watch recording
Presenters: Todd Havekost and Brent Phillips
Db2 produces extensive SMF metrics that can provide great insights into how the many facets of your Db2 environment are operating.
This webinar will include how to:
Request link to watch recording
Presenter: Todd Havekost
In this session on gaining insights from the Db2 SMF data series, we take a deep dive into the Db2 Accounting (SMF 101) data.
Our learning journey will include:
Request link to watch recording
Presenters: Walter Auerochs (DATEV), Hendrik Klimek (DATEV), and Todd Havekost
In this session, the third in our Db2 webinar series, two users at one of our customers share how they have leveraged IntelliMagic Vision to help manage their environment more effectively and enhance Db2 availability and performance.
They will present several examples of how dynamic navigation through their SMF data has helped them quickly identify and solve business problems. Their story also shows how a common, intuitive interface providing visibility into data types across the platform is enhancing collaboration among their infrastructure teams.
Request link to watch recording
Presenters: Todd Havekost
CICS is the leading z/OS transaction manager. Watch this session and find out how much you can learn through exploring CICS Transaction (SMF 110.1) and CICS Statistics (110.2) data.
Our learning journey will include:
Request link to watch recording
Presenters: Todd Havekost
MQ Statistics and Accounting SMF data provides extremely useful insights into MQ management and operation. Unfortunately, extracting those insights from the SMF 115 and SMF 116 record types is challenging – and raw data pulls lack context or accessibility.
With proper access to the insights derived from this data, analysts can optimize their z/OS MQ configurations and activity more effectively and efficiently and have better visibility and control over their environment. This webinar demonstrates how to overcome these common challenges and offers a guide to analysts responsible for and working with MQ environments.
Request link to watch recording
Presenters: Todd Havekost
IBM’s introduction and active promotion of the Software Consumption (SC) solution within Tailored Fit Pricing (TFP) brings with it a fundamental change in how sites will approach managing and reducing software expense.
Instead of being driven only by CPU usage during the monthly peak four-hour interval, under the SC model, software is charged based on the total CPU consumption for the entire month (at a much lower unit price). This new SC model brings with it a new set of challenges for managing an environment and at the same time potentially significant opportunities to reduce software expense over the long term.
Request link to watch recording
Presenter: John Baker
Duration. Seems simple enough: how long do I give a transaction to get done? The reality is somewhat more complex. Firstly, the duration is not wall-clock time, it’s weighted Service Units. And what are your ‘weights’ (coefficients) anyway?
Multiple periods are one of the most powerful – and dangerous – tools in the performance analyst’s arsenal. They can ensure critical work gets done and runaway work can’t take over. This webinar provides a timely update to setting up or reviewing this critical area.
Request link to watch recording
Presenter: John Baker
As of z/OS 2.5, WLM coefficient settings were removed; IBM made a good move in sunsetting an obsolete set of variables. I/O priority should follow. WLM I/O Priority has been around for a long time and is in use at many data centers. A recent survey revealed many sites are not even clear on what the setting does. While the default is ‘off’ the recommendation in the WLM Planning manual is ‘on’. In this presentation, we will demonstrate why this option should be turned off and perhaps removed entirely like the coefficients.
Request link to watch recording
Presenters: Jerry Street and Sally Mir
DB2 Buffer Pool performance is essential for applications in a z/OS environment. Buffer pools in data sharing configurations have corresponding coupling facility structures, so monitoring them from a coupling facility perspective as well as a DB2 perspective is key.
This zAcademy session covers the basics of coupling facility performance and dives into DB2 performance, specifically DB2 group buffer pool performance.
Request link to watch recording
Presenters: Todd Havekost
SMF data has the potential to provide great insights into the operation of components across the z/OS platform, but infrastructure teams face many challenges seeking to maximize the value they derive from that data so that they can deliver high levels of availability and performance.
This webinar will explore how SMF data analysis becomes far more effective when a common, intuitive interface provides integrated visibility into the many types of SMF data produced by components across the z/OS platform.
Request link to watch recording
Presenters: Joe Hyde
The zHyperLink Express feature was announced by IBM in September 2017, providing up to 10x better response times in read cache hits. The roll out to production has been slower than most new interface technologies on the mainframe given that it’s an adjunct and not a replacement to FICON.
Join Joe Hyde as he shares some recent DB2 production data which highlights the speed of zHyperLink read I/Os.
Request link to watch recording
Presenters: Jack Opgenorth and Todd Havekost
Given the importance of security in today’s world, managing network encryption is a critical requirement for almost every z/OS environment. zERT data can be of great benefit to network security teams who are responsible to ensure that transmitted data is encrypted in accordance with corporate standards.
Join this session to learn more about the value provided by zERT and other SMF data.
Request link to watch recording
Presenters: Dan Speer and John Baker
Comprehensive health checks for your z/OS mainframe are different than day-to-day performance analysis and management and are essential for heading off budding availability risks before they grow out of hand.
We’ll cover real-life case studies that assisted clients in avoiding potential performance or availability issues and examples of cost savings opportunities identified during a mainframe health check. Even analysts with the finest tuned mainframes will walk away with some new ideas.
Request link to watch recording
Presenters: Sally Mir and Todd Havekost
Db2 Accounting data can provide a very rich source of insights into your workloads and performance. Sites often face significant barriers to unleashing the value of that data, including the ability to focus analysis on a desired subset of data despite overwhelming data volumes.
This recording will present several real-life examples where good visibility into Accounting data enables issues to be detected and resolved or insights to be derived.
Request link to watch recording
Presenters: Gabe Tully and John Baker
In today’s age of ‘app’ proliferation and data overload in most areas of z/OS infrastructure performance, one would think we would have access to endless statistics on application behavior. While this is largely true in the distributed space, it remains an elusive goal in our beloved mainframe.
In this session, John Baker and Gabe Tully try to break down the barriers between z/OS and distributed systems and communicate specific methods both sides can use to classify workloads properly.
Request link to watch recording
Presenters: Frank Kyne and Todd Havekost
XCF Transport Class simplification functionality was implemented in z/OS 2.4, with an objective to eliminate the need for separately managed transport classes at most sites.
You can expect to come away with a better understanding of how XCF operates in general, tips about how to optimize their environment for the new paradigm, and information about what can be learned from the new Path Usage metrics.
Request link to watch recording
Presenters: Dr. Gilbert Houtekamer and John Baker
The CPU Measurement Facility (CPU MF) was introduced with the z10 and contains a wealth of detailed information about processor performance.
Dr. Gilbert Houtekamer took notice of the substantial processor cache design changes in the z16. They are sometimes seen as magic numbers, but it is important to gain insight into how they will impact the operation and efficiency of your workloads.
Request link to watch recording
Presenters: Todd Havekost
CICS produces extensive SMF metrics that offer rich potential insights for its management and operation. But sites often face significant barriers to realizing the potential value of this data.
This webinar will introduce you to several best practices for enhancing your analysis of CICS Transaction (SMF 110.1) and CICS Statistics (110.2) data.
Request link to watch recording
Presenters: Todd Havekost
Most mainframe sites are regularly on the lookout for potential ways to reduce expenses. The adoption by many sites of TFP-based software license models where all CPU is chargeable has made this an even higher priority.
This webinar expands your awareness of ways sites have reduced CPU and expenses so that you can identify ones potentially applicable in your environment.