Tuning for Improved Performance and Lower TCO:

Document ID : KB000049807
Last Modified Date : 14/02/2018
Show Technical Document Details

Description:

Steps to make optimal use of your computer resources while maximizing CA IDMS performance include:

Solution:

Tuning for Improved Performance and Lower TCO:

Steps to make optimal use of your computer resources while maximizing CA IDMS performance include:

  • Use of the IBM z Integrated Information Processor (zIIP): By exploiting zIIP, CA IDMS can help you deliver more computing capacity and throughput without additional system hardware resources. The zIIP (IBM System z Integrated Information Processor) is a specialty mainframe processor designed to help free general computing capacity from the Central Processor (CP). Use of the zIIP feature can help you lower your TCO (total cost of ownership) on the mainframe.

  • Use of the High Performance Storage Protection Option: HPSPO protects the integrity of your systems by preventing user programs from overwriting CA IDMS and operating system storage, from overwriting CA IDMS and operating system code, and from executing privileged instructions. With HPSPO you do not have to make the difficult choice between system integrity and CPU. You can implement storage protection with minimal CPU cost. The use of full storage protection incurs significant CPU time.

  • Use of the scratch in memory and expandable scratch features: Enabling the use of scratch in memory significantly improves CA IDMS runtime performance by eliminating file I/O to a scratch area. Defining scratch in memory also improves DBA productivity by eliminating the need to define and maintain a scratch area for every CA IDMS system. Enabling extensible scratch reduces the possibility of task failures due to insufficient scratch space at runtime.

  • Use of recommended settings for system generation parameters: Adjusting the CA IDMS system generation parameters appropriately for your site will provide optimal runtime performance of a CA IDMS system. This will help to meet Service Level Agreements for response time and effectively use available computing resources.

  • Use of CICS threadsafe applications: Using CICS threadsafe applications can significantly increase throughput of tasks in the CICS region, improve response time and decrease costs. The CICS TS Open Transaction Environment (OTE) enables multiple tasks to execute the same THREADSAFE program simultaneously operating on different TCBs.

  • Use of the type 4 JDBC driver for Java applications: The CA IDMS Server JDBC driver is tightly integrated with CA IDMS SQL processing for optimized SQL access to both non-SQL and SQL-defined CA IDMS databases. The type 4 JDBC driver provides for optimal performance and response time for web and client/server Java applications accessing CA IDMS. With the type 4 JDBC driver, the Java client program is able to communicate directly with the CA IDMS region through native TCP/IP. This eliminates the need for any middleware address space, such as CCI, and frees mainframe processing for other workloads.

  • Use of recommended CICS TS settings: Use the SYSCTL DDNAME parameter of the CICSOPT macro instead of the SVC and CVNUM parameters to specify the CA IDMS system to access. If same CICS region is used to access multiple CA IDMS backend systems, then a separate IDMSINTC interface module must be created for each unique CA IDMS central version. By using the SYSCTL DDNAME parameter of the CICSOPT macro to specify the CA IDMS system, the name of the CA IDMS system to be accessed can be controlled at runtime. This provides for flexibility and enables a customer to respond quickly to changes in system workloads by bringing up additional CA IDMS back-end systems.

For a more detailed discussion on each of the above items, please see the CA IDMS Version 18.0 Best Practices Guide:
https://support.ca.com/cadocs/7/CA%20IDMS%20DB%20Version%2018%200%2000-ENU/Bookshelf_Files/HTML/IDMS_Best_Practices_ENU/index.htm