zWorkload Reporter and zGUARD provides insight and real-time automation to optimize performance and lower IBM’s Monthly License Charges (MLC).
Understanding and controlling the MLC cost is a daunting challenge that IBM Z mainframe customers face. IBM Z customers can easily find themselves in a situation where they exceed the Defined Capacity (DC) on one LPAR, resulting in a surcharge, while they have headroom to spare on other LPARs. The trick is to stay on top of this by constantly monitoring workloads and shifting activity, workloads and resources among LPARs to ensure they don’t exceed the defined cap (DC) and that applications performance is not derogated.
This requires a skilled mainframe staffer with both a high level of z/OS skill and familiarity with workloads and LPARs. While you’re at it throw in knowledge of data center operations and the organization’s overall business direction. Finding such an expert is costly and not easily spared for constant monitoring. It’s a task that lends itself to automation.
And that’s exactly what, zGUARD and zWorkload Reporter addresses for IBM Z mainframe customers. Our solutions analyze and report on workloads, then makes dynamic adjustments reducing Monthly Licenses Charges (MLC) without affecting application performance. On average companies that actively manage and effectively prioritize their mainframe workloads save 10-15 percent more on their monthly license charges than those who use a more passive approach. The savings, can add up fast. For example, one customer taking this approach cited saving 161 MSUs, which translated for that organization to over $55k that month. Given that a mainframe shop spends anywhere from few hundred thousand to millions of dollars per month on MLC charges, savings of just a few percent can be significant.
zWorkload Reporter is a cloud web reporting service that provides insight that helps drive WLM configuration decisions, analyze problems and offers an intuitive and dynamic, web-based dashboards which enables mainframe teams to collaborate with confidence, quickly answer questions, resolve issues, and identify opportunities for improvement.
zWorkload Reporter is offered as a cloud enabled solution. The zWorkload SMF tool extracts the required meta data as part of the nightly SMF process and automatically uploads the data to the zWorkload reporter cloud, where it is analyzed, correlated and displayed in pre-defined customizable dashboards and reports. zWR is installed on servers in our secure data center or at one of our partner locations.
zGUARD © is an on-premise software solution that dynamically automates the optimization IBM Z mainframe MSU distribution, reducing Monthly License Charges (MLC) without impacting workloads. zGUARD is an open programable solution, that supports scripting in a wide array of programming languages. Scripts can be used to trigger actions based on event based alerting in real-time. Actions are flexible and adaptable, they can control/distribute batch processing, change Defined and Group capacity values, monitor and control performance by optimizing consumption intervals.
Depending on which effect the customer wants to achieve change can be based on limits or system-settings with zGUARD ©. The adaptions can happen every minute or as necessary. On the following charts, notice how the R4HA (Rolling 4 Hour Average) blue line with white border changes as the adapted Defined-Capacity value dynamically optimizes MSU distribution.
Customer problem: Customer was experiencing rising MLC (Monthly License Charges) and tried to manually manage the actual MSU usage of individual workloads to lower MLC cost was not working. The manual efforts attempted were time-consuming, frustrating and unreliable.
Requirement: Customer wanted an automated solution that could be adaptable to his specific needs and ever-changing workloads requirements. Customer wanted a lower DC when the working day started and then use a dynamically changing DC during lunch time and all scenarios needed to avoid unnecessary peaks and reduce MLC cost.
Solution: zWorkload Reporter was intially used to understand workload requirements and formulate an optimization plan, then zGUARD was implemented on site to automate specifc changes to DC to support all scenarios and goals customers required. Today both soltuions are continously proactively monitoring, reporting and adapting to ensure customer goals are maintained as enviroment changes and new requirements arise.
We influence the consumption of the LPAR with a dynamically changing defined capacity. We cap the system/LPAR only when necessary and only when it saves money. Without capping the system during peak hours the consumption would be higher and therefor the 4h rolling average. The 40 MSU is an estimate if the system had no defined capacity at all. Then the 4h rolling average would have been higher (by 40 MSU probably). In fact this means that an LPAR is squeezed down to a minimum MSU value unless time critical workload starts suffering for CPU (MSU).
Customer achievement: ~40 MSU – zGuard capped the system/LPAR only when necessary and only when it saved money. zGuard understood which and how much “low priority” workload were in the system and decided upon that measurement, who to cap and when to reduce the R4HA ( Rolling 4 Hour Average); saving the customer thousands of dollars on their MLC (Monthly License Charges) without impacting workloads.
Problem: Spikes in R4HA (Rolling 4 Hour Average) due to high MSU consumption peaks driving up cost. The customer had a lot of batch processing during night and day. Batch tends to use all CPU (MSU) which it can get.
Requirement: Customer wants to keep the R4HA (Rolling 4 Hour Average) as flat as possible while guaranteeing maximum throughput. No fixed Defined Capacity during the month.
Solution: zGUARD understands what is available (analyzes LPARs usage) and distributes only what was used so far and what was charged for already ($). Example: SCRT starts 2nd of month, 10th will use 50 MSU, from now on zGuard tries to run the system with 50 MSU. If new peak is reached on 13th for example of 60 MSU, then the same game applies for 60 MSU.
Customer achievement: ~15 MSU - With zGUARD automating DC capping the R4HA was reduced, resulting with customer saving thousands of dollars on their MLC (Monthly License Charges) and where workloads were not impacted.
• zWorkload Reporter offers a user-friendly and in-depth cloud web reporting service to track activity and quickly identify opportunities for improvement.
The combination of task automation and detailed activity monitoring frees precious time for your mainframe experts to focus on other high value-added activities. In addition, our expert services offered by SVA Software and our certified partners ensure WLM settings are always optimized.