It’s Monday morning. The IT operations staff of a large company fires up a critical mainframe application and observes that the application is consuming only a fraction of its normal computing capacity. The staff assumes the application is down and begins to panic.
After several anxious calls, the staff discovers that a DBA had analyzed the application and found that two DB2 transactions accounted for most of the resource consumption. The DBA tuned the offending SQL statements over the weekend, which lowered consumption from about 30 percent to 2 percent, resulting in a substantial cost reduction.
This scenario actually occurred, and it illustrates how optimizing mainframe resource consumption can help IT cut costs.
Mainframe costs are linked to capacity utilization. So reducing resource consumption plays an important role in optimizing mainframe spending. The key lies in understanding how you’re paying for IBM and ISV software license charges. Many of these contracts are linked to peak capacity or to a rolling four-hour average of consumed processing capacity. Reducing your CPU/MIPS/MSU consumption can often have an immediate effect on lowering costs. Mainframe management technology can help make this happen.
Here are four factors to consider in optimizing costs and selecting tools to help you:
1. Analyze and tune. IT managers sometimes assume that all processing inefficiencies have been squeezed out of the mainframe over its many decades as a data center workhorse. If you make that same assumption, you may overlook opportunities for improvement. For example, many mainframe computing sites outsource application development to keep the IT staff small while still meeting business needs. The applications, however, aren’t tested in the production environment. When they hit the large-scale transaction volume of the actual production environment, they may run more slowly or consume more MIPS than expected. To avoid this situation, analyze applications with an eye toward reducing MIPS consumption. Tools can help you identify performance bottlenecks and inefficient code, such as poorly performing SQL statements. Performance optimization tools, such as DB2 tuning tools, can be used to address the problems you find and reduce peak capacity requirements.
2. Eliminate surprises. Most cell phone providers now warn customers when they’re approaching their plan’s limits. In a similar manner, mainframe monitoring tools can monitor capacity and alert you when consumption is approaching the limits you’ve set. As a result, you know immediately when you need to head off a major spike in monthly costs caused by a runaway application.
3. zIIP it wherever possible. Moving appropriate processing tasks to System z Integrated Information Processors (zIIPs) offers a cost-effective way to lower peak capacity requirements. Some tools, for example, automatically shift work to zIIPs, thereby freeing up mainframe capacity.
When evaluating tools for monitoring and managing mainframe capacity utilization, look for ones that are efficient in their own use of mainframe capacity. Seek out tools that offload a substantial portion of their processing to zIIPs. Tools that meet these criteria won’t cancel out your efforts to reduce peak mainframe capacity consumption.
4. Test drive first. In evaluating tools, assess in advance their potential to lower your peak capacity. For example, look for a vendor that can test drive tools that analyze your System Management Facility (SMF) data to determine your peak capacity and identify what’s running during peak periods. You can leverage that information to determine the return on investment you can expect from the tools. You may discover that by simply moving a particular batch job to a different window, you can reduce your peak capacity and monthly cost.
Many organizations are already benefitting from mainframe cost optimization. The manufacturer cited earlier reduced transaction costs by about 100 percent for a savings of nearly $350,000 in the following year. A large bank shrank the running time of a job from one hour to one minute. The bank now runs its business using only about one-fourth to one-half the mainframe capacity required by other banks of similar size. By applying these tips, you can bring similar benefits to your enterprise.