How to Hot-Rod Your Mainframe Applications and Save CPU
In a commissioned study conducted by Forrester Consulting on behalf of Compuware in November 2016, researchers determined “96 percent of companies [involve] the mainframe for new business initiatives.”
On the other hand, “90 percent of IT professionals reported that their mainframe organizations are experiencing challenges with application development and delivery.”
The good news is digital business is driving activity on the mainframe, ensuring the platform’s longevity. The bad news is many business-critical mainframe applications are struggling to meet new performance levels that increased activity demands.
Tweaking the Speed Limiters on Your Apps
How fast your car can go or how much gas it uses depends on what kind of car it is and how its computer has been programed. The computer can limit how fast your car will go, how fast it will accelerate and what kind of fuel economy it will get. This is all determined by the manufacturer of the vehicle.
Let’s say your car can accelerate from zero to 60 in 10 seconds with the computer that your car’s engine has. This is the speed you’re used to—it’s how you expect your car to perform. But if you know how to hot-rod your car, such as by tweaking your computer, you may be able to make your car accelerate from zero to 60 in eight seconds, hypothetically, at the cost of some fuel economy.
Developers often see applications from this perspective as well. Applications perform how they’re expected to or were originally coded to perform. They run as long as and use as much CPU as they’re expected to. Unless that application performance noticeably degrades, it’s unlikely developers will feel the need to go look at a program and begin pinpointing inefficiencies.
But like a car, developers can look under the hood to see what’s going on and start tweaking applications to make them perform better. This is something that should already be happening for the sake of application performance, but more importantly, these applications now need to perform at peak levels to cope with the increased mainframe activity digital business drives.
When your applications perform better, you’ll save CPU, save money and have more funds to invest in improving your mainframe or some other portion of your business. Ultimately, this leads to your enterprise becoming more competitive.
Finding Application Inefficiencies
To make your car go faster, you need to know where to locate the car’s computer, how to interface with it and, finally, how to reprogram it. To make your applications perform better, you need to know where inefficiencies are occurring that damage performance and cause spikes in CPU usage. This requires a state-of-the-art application performance monitoring and analysis tool that can pinpoint application performance defects causing excessive CPU consumption.
That’s why our partner CPT Global—a worldwide IT consulting services firm specializing in capacity planning, performance tuning and testing—uses Compuware Strobe as its preferred mainframe application performance monitoring and analysis tool to help its customers save CPU. But why use Strobe over any other application performance analysis and monitoring solution?
Learn what makes Strobe transcends other tools in “Using Compuware Strobe to Save CPU: 4 Real-life Cases from the Files of CPT Global.” During this webcast, we go through four real-life cases of how CPT Global has used Strobe to help its customers save CPU by identifying inefficiencies in:
- COBOL code
- VSAM access
- A vendor product
- CICS system settings
Finding inefficiencies in your applications to save CPU is something you should already be doing, especially when the mainframe is experiencing even more activity from digital engagement. When you save CPU, you save money you can invest elsewhere, and that makes you more competitive in an economy where a two-second difference of acceleration from zero to 60 truly matters.
Latest posts by Spencer Hallman (see all)
- Where Was the Mainframe at Big Blue’s InterConnect 2017? - March 27, 2017
- How to Hot-Rod Your Mainframe Applications and Save CPU - December 8, 2016
- The CICSPlex Starburst Effect on Application Performance Measurement - June 21, 2016