Blog > August 2020 > How to Maximize Mainframe Performance and Flatline Capacity Costs

How to Maximize Mainframe Performance and Flatline Capacity Costs

For many enterprises, the mainframe is still the mission-critical backend system for business logic and transaction processing that drives digital transformation. Several industry surveys have found that mainframe workloads are growing year over year as new business initiatives like DevOps, automation and modernization extend application development and delivery across distributed, cloud and mainframe systems.

For years, evolving customer expectations and business demands have fueled the need for cross-platform application growth and increased computing capacity. Not only have IT organizations had to manage this growing complexity while being asked to reduce datacenter costs, but new challenges like the coronavirus pandemic have forced businesses to operate on even leaner budgets without retreating on digital transformation.

As workloads grow and businesses push harder to innovate, choosing to reduce mainframe capacity to cut costs will only lead to decreased application performance and dissatisfied customers. At the same time, paying for more MIPS (millions of instructions per second) to increase capacity on the mainframe and support application growth is a burden most IT budgets can’t afford to bear, let alone during pandemic times.

Fortunately, IBM continues to optimize the IBM Z mainframe to increase its operational efficiency, in tandem with new cloud-like pricing models that ease costs. Even so, as more enterprises increase mainframe workloads, there remains a need to mitigate capacity pricing increases by maximizing the utilization of existing mainframe resources.

To accomplish this, IT teams need mainframe performance and capacity management tools to understand how existing capacity is being used and where adjustments are needed to enable application and transaction growth without having to purchase more capacity.


Improve IBM Z performance with ASG


How Mainframe Performance Tools Help Maximize Existing CPU MIPS

A critical step towards maximizing the use of existing mainframe resources is to implement a good mainframe performance and capacity management solution that provides a visualized and data-driven approach to reducing or controlling MIPS consumption. When it comes to choosing the right tool, look for a solution that enhances your ability to:

  • Identify which workloads are causing high CPU consumption and when, and what is driving CPU spikes that constrain processors and have adverse effects on mainframe performance and budget
  • Forecast future MIPS consumption per workload over time, based on projected application growth and changing business requirements - exposing whether there is enough existing capacity to address application growth without purchasing more processing power
  • Analyze the applicability and potential benefits of using less expensive specialty processors such as zIIPs to process the same work and reduce MIPS consumption - reserving more expensive MIPS for critical workloads that must run on general processors  
  • Determine the optimum configuration settings for various logical partitions (LPAR) and determine the effect of potential changes to settings for effective distribution of CPU resources
  • Model and chart the placement and movement of LPARs on different CPCs to maximize resource utilization and control MIPS growth
  • Analyze the effectiveness of using defined capacity for z/OS LPARs to control MIPS consumption and the possible adverse effects of soft capping, which reduces an LPAR’s access to CPU resources
  • Accurately identify delays when an LPAR is accessing processor resources, versus actual high CPU MIPS consumption, to avoid unnecessarily increasing MIPS capacity and to optimize configuration options and tuning instead.

Key Benefits from Your Mainframe Performance Management Solution

A modern mainframe performance management and capacity planning solution will enable you to accomplish advanced tasks with more efficiency and aid your business and digital transformation efforts with the following benefits: 

  • Reduced CPU consumption and recovered processing time
  • Identification of system, database, I/O and application performance inefficiencies
  • Pinpointing of application execution delays and execution metrics
  • Understanding of performance profiles at the transaction level
  • Improved application throughput and planning for growth to avoid costly upgrade charges

Collecting both physical and virtual performance and capacity data across your enterprise and analyzing it automatically will make it easier to know where your problems are, where you can improve performance and how to assess what needs to be done to improve your overall IT utilization.


Get Started with ASG



millions of instructions per second