You run CICS applications because they perform well, and have been reliable from the start. Unfortunately, with increasing demands on your systems, you still need more performance from your key assets—but your costs are rising while your mainframe IT budget is not. Performance requirements, continuous availability as well as expensive CPU cycles, are daily concerns for your business, finance and IT management groups. CICS application performance suffers from some common challenges: the amount of data to process is growing steadily (caused in part by increased mobile transactions accessing your CICS applications), straining system capacity and perpetually increasing mainframe operating costs. DataKinetics CICS optimization solutions are many and varied, but are delivered by just two powerful products: tableBASE Data Access Accelerator, and VTS Edge.
Speed up CICS processing and reduce OPEX using in-memory technology
Generally, data access follows the 80-20 rule— 80% of your data is accessed only 20% of the time, and 20% of your data is accessed 80% of the time. The 20% of your data that you are accessing 80% of the time is the reference data that you access many times for every business transaction. This is typically where most costs get eaten up, and represents a significant opportunity to increase performance and decrease operating costs—considerably. DataKinetics tableBASE can be used to do just this. Copying some of the most often accessed data into tableBASE optimized in-memory tables allows your CICS applications to access that data up to 100 times faster than is possible using any other method.
The image below shows how the data that is used most often during processing is accessed from tableBASE optimized in-memory tables; while the rest of the data is accessed from DASD using current techniques. In this way, CICS applications can run much faster—and the difference can be astounding.
As part of transaction processing, temporary files are often created for the purpose of sorting, Using temporary access-optimized in-memory tables instead can save CPU resource usage and processing time for every business transaction.
Direct benefits of using tableBASE with CICS applications
In CICS transaction processing environments, customers have seen as much as 90% reduction in processing run time for their applications in areas associated with data access and inter-region communications. They also use up to 65% less CPU and up to 80% less I/O resource usage, sharply reducing the operational expense burn rate. In some cases, the savings can be so dramatic that planned capital expense spend can be deferred, or even eliminated. This is why tableBASE is the tool of choice for many Fortune 500 companies for improving mainframe and CICS application performance, reducing resource usage and reducing operational costs of their critical assets.
Maximize zIIP Processor usage
Normal zIIP operation is limited by the need to access data using I/O. Such activity cannot be performed on the zIIP— an application must switch to the Central Processor for I/O activity. If it were possible to access data while running on the zIIP, organizations would be able to save considerably on operational expense. While it is not possible to access data from zIIP using I/O, there is no reason that you cannot access data that is resident in memory. In fact, data contained in DB2 buffers can be accessed while processing on the zIIP, but that only helps a small amount. To use the zIIP processor for data that you access most often, you need a data integrator like VTS Edge.
Direct benefits of using VTS Edge with CICS
VTS Edge data integration allows data access from high-performance in-memory tables that are zIIP eligible. For best results, you will identify the data that is read most often by your applications (typically a very small amount), copy that data into the VTS Edge shared in-memory tables, and access the data from there using an API. This is shown in the image below.
The benefits are two-fold: the data can be accessed much faster than is possible from buffered DASD, and it is all zIIP eligible. Since the in-memory data is ideally the data that you access most often, the cost required to access this data will be reduced significantly.
Your CICS applications will still access data from other sources off the zIIP, but ideally, that will be minimized—limited to occasionally read data, data that is updated occasionally—data that will have a lower impact on the cost of your processing.
If the CICS application accesses data from the in-memory tables much more often than it does from I/O, the overall usage of cheaper zIIP Processor can be maximized, and the operational expense of running the application can be dramatically reduced. A further benefit is that you effectively maximize the availability of General Processors (GP) for other important tasks. Savings can be much greater if multiple CICS applications (or other applications) can be zIIP optimized in this way