Worldwide economic uncertainty in recent years has put significant pressure on CIOs to at least keep IT costs level and more likely push for cost reduction across the board. Gone are the days when performance issues were relatively easily handled by adding or upgrading hardware and such purchases were part of the routine budget cycle – today everyone is expected to “sweat the assets” and “do more with less”.
One of the most effective ways of reducing or containing mainframe costs is through better management of CPU consumption. By slowing down growth of CPU usage and managing workload placement, hardware and software upgrades can be deferred thereby allowing organisations to keep costs down and profitability up.
Reasons for MIPS Growth
The main reasons for growth can be categorised as follows:
- Increased transaction rates. Whilst it is generally good news for businesses to see transaction rates on the rise, with z/OS usage based pricing this increase in workload can push up software costs and can also negatively impact application performance.
- More demanding applications. As enterprises strive to remain competitive, more business operations are being automated and packaged business applications from vendors such as SAP and Oracle are becoming more common. Such applications are highly capable and flexible but typically require more computing resources than their bespoke equivalents. Other factors can also result in increased application CPU usage, such as the move to Java-based applications and reductions in development and testing time due to business and cost pressures.
- Middleware overheads. IBM is constantly enhancing critical enterprise middleware components such as DB2 and CICS, adding new capabilities to improve productivity and reduce development and operational costs. However, such enhancements can often entail additional CPU overheads and many organisations find themselves paying the CPU cost for the new function without always being able to exploit the benefits.
The Effects of MIPS Growth
The growth factors discussed above can have a direct impact on the cost and performance of an organisation’s mainframe applications:
Cost – The major driver for many IT teams is the need to reduce mainframe resource usage and thereby potentially defer hardware upgrades and reduce monthly MIPS costs. There are also human costs to consider: maintaining an underperforming system takes more time and resource from IT teams, and adds pressure from the business teams who are calling for improved response times.
Performance – Typically, any significant increase in the amount of CPU used by a given workload will result in an associated increase in transaction elapsed times. For performance-critical online workloads, that increase can translate directly into poorer critical business metrics such as customer satisfaction and retention.
Just throwing more MIPS at a poorly-performing workload does not always address the issue. A 2-hour response time may be reduced to 1.5 hours with more CPU time being available, but the problem might be due to a poor access path and some expert DBA attention could get it down to 5 seconds. This is especially true of application performance tuning, which is where the majority of performance issues tend to lie.
By analysing your current environment and identifying opportunities to optimise and tune your mainframe workloads you can regain control of spiralling costs.
The key to a successful workload tuning exercise is to know where you’re starting from. It is vital to understand exactly where workload peaks occur, and how they contribute to software licence costs, before you can really be effective with your workload tuning. I have seen many large mainframe customers struggle to get a clear view on when their workload peaks occur across all mainframe workloads. Only once these peaks have been identified can organisations really bring down the cost of their mainframe software licencing through tuning activities.
One of the major challenges in any environment, but particularly with client/server applications, is determining which component is responsible for poor response times, although the tools for this are improving. Another related challenge is “skills silos”, where a client has the individual skills necessary to resolve a particular issue but no single person has the whole picture and internal culture and politics prevent the individuals from communicating and collaborating effectively.
Regain Control of Software Charges
The majority of mainframe users have significant potential for reducing resource consumption (and therefore costs) through performance tuning of key workloads. This is especially true for those with older applications that haven’t been actively maintained for a while or who have lost some of their deep middleware skills through retirement or redundancy. By tuning these workloads, ongoing software costs can be reduced and mainframe upgrades potentially deferred. In addition, application performance will be enhanced and overall Total Cost of Ownership (TCO) reduced.
Through the effective analysis and tuning of workload peaks it is typically possible for organisations to reduce their CPU charges by at least 3-5%. For organisations running very large workloads this can equate to savings of tens of thousands of pounds per month.
The benefits of mainframe workload tuning can be felt across the entire business. From the CFO who will see significant reduction in IT spend through to the IT teams who benefit from improved application performance and thus improved customer service, a thorough tuning exercise can indeed improve many aspects of business performance.
Find out more about our Mainframe Cost Reduction Service