Blog

10 mainframe performance and capacity questions to ask yourself in 2023

The beginning of a new year is a good time to stop up and ask questions. Here are a few questions about capacity and performance we recommend thinking about in 2023.

Does the rest of the company understand how their behavior affects performance and capacity? How can you create (or improve the existing) mapping of transactions and jobs (including DDF) to applications and business areas. Creating self-service reports to users within development, finance, and the business lets them help in optimizing IT infrastructure capacity to save costs.
Will you be changing your primary software licensing principles in 2023 (for example from WLC to TFP)? Is your reporting system geared for the change? Your charge-back? Adding MSUs over time seems easy enough, but who will pay for any additional hardware capacity necessary to support your peak load? Does the change match your old capping definitions (or lack thereof)? Will the new billing model introduce new cost drivers you should be focusing on?
What new technologies should you be investigating by reading up on them or checking out webinars? You may want to know more about the z16 (which is yet an entirely different beast), for example, in time before upgrading to one. At least you should make sure to measure and analyze before/after scenarios to ensure that the new technology is living up to your expectations.
Will you be using more data-in-memory during 2023? How much additional Central Storage will you require?
Is it time to re-visit your Sysplex setup? How many MSUs are you using in spin-loops while your Systems are waiting for synchronous Coupling Facility requests to finish? Do you really need to share all your data everywhere or can you reduce your Sysplex overhead?
Do your LPAR configurations need to be adjusted? Are you getting a lot of ‘low’ vertical affinity engines? What about zIIP eligible workload running on CP’s? Are you using all the logical CPUs on the LPAR, or has hyper dispatch parked some of them most of the time? Do you need to adjust the LPAR weights?
How have the business volumes changed during 2022, and what are the expectations for 2023? How do you relate business volumes to capacity requirements and translate those into specific physical or logical configuration changes?
Have your company’s business priorities changed? Do these changes need to be reflected in the capacity made available to different applications? What about Workload Manager? Do you need to adjust the importance or goals of any of the workloads?
What are your primary capacity-related cost drivers? Are the primary cost drivers closely related to your core business? Or do you have ‘rouge’ workloads – that is unpredictable, low business value workloads that use lots of capacity at inconvenient times?
What has changed during 2022 and how do you plan to manage changing capacity needs in 2023? Are there new applications, new usage patterns, new technologies, new configurations that impact your capacity and performance? How do you identify them before they become a problem and evaluate the impact after they have been rolled out?

The answers to all these questions require good data, good tools to understand the data, and perhaps having a trusted advisor to help you understand the implications. Give us a call in the new year if you want to sit down and talk about how we can help you with this.