UNICOM PIE/CICS-NonStop

 

PIE®/CICS-NonStop

Keep CICS Running with Fully Initialized Backup Regions

 

OVERVIEW

PIE/CICS-NonStop is an optional PIE/CICS component and runs multiple, fully initialized regions. Backup regions are available immediately when problems occur. PIE/CICS-NonStop re-assigns users while recovery procedures continue in the failed region. PIE/CICS-NonStop does not have to wait for the entire period to complete a dump. Usually, a dump takes two minutes or less. In most cases, users are re-assigned to the backup region immediately after a failure. PIE/CICS-NonStop handles both local and remote terminals. Unlike XRF, PIE/CICS-NonStop does not need to continuously monitor CICS regions. Accordingly, the performance overhead of running PIE/CICS-NonStop is less.

PIE/CICS-NonStop automatically routes users to an appropriate region. Users need only one logon procedure, regardless of how many regions they use. No special knowledge or training is required to access different regions. PIE/CICS-NonStop automatically routes users to different regions to distribute user load across regions exactly to your specifications. This method of user distribution is called load balancing.

CICS availability is critical. The cost of even short outages can be extremely high. Outages are the inevitable result of software or hardware problems, and required periodic maintenance. Minimizing the number of outages and the length of system down time is a critical issue at most sites.

Recognizing the need for recovery systems, you could use an automated operator to restart a region after it fails. The time needed to restart a region can range from a minute or more. If the system produces a dump, you must wait for the dump to complete. In addition, you must wait for all CICS recovery procedures to finish. Your system is unavailable for most of the recovery period.

Most sites cannot have mission-critical applications unavailable for the entire recovery period. In response, IBM created the Extended Recovery Facility (XRF). With XRF you have two regions: a regular region intended for normal CICS transaction work and a backup region. In the event that the regular region fails, XRF automatically re-assigns users to the backup region.

Even with XRF, the backup region is only partially initialized. It usually takes several minutes to fully initialize the region. In addition, you still have to wait until the dump and the recovery procedures are completed. Also, XRF can only reroute remote terminals. Local terminals are left hanging.

PIE/CICS-NonStop provides a better solution. PIE/CICS-NonStop runs multiple, fully initialized regions. Backup regions are available immediately when problems occur. PIE/CICS-NonStop re-assigns users while recovery procedures continue in the failed region.

PIE/CICS-NonStop does not have to wait for the entire period to complete a dump. Usually, a dump takes two minutes or less. In most cases, users are re-assigned to the backup region immediately after a failure. PIE/CICS-NonStop handles both local and remote terminals. Unlike XRF, PIE/CICS-NonStop does not need to continuously monitor CICS regions. Accordingly, the performance overhead of running PIE/CICS-NonStop is less.

At a Glance

• Auto route users across low use CICS regions
• Users only need a single logon proc
• No waiting or delay during CICS dump
• Supports local and remote terminals