7.5 Base Enhancements – IBM IPP Wiki

IBM PowerHA SystemMirror for IBM i 7.5 with HA version 5.0.1 brings several enhancements to the PowerHA portfolio, including:

A simplified single edition offering

Reduced data transfer during resynchronization with Geographic Mirroring Compression

Reduction in Time Required to Perform CRG Switchover for Selected Configurations

Simplified Single Edition Offering 7.5 HA 5.0.1

The PowerHA SystemMirror for i offering is simplified to have a single edition containing all functionality.

Reduced Data Transfer with Geographic Mirroring Compression 7.5 HA 5.1.2 7.4 HA 4.7.2

Geographic mirroring performs disk-level replication. This means that all changes on disk in an Independent Auxiliary Storage Pool (IASP) are replicated over the TCP/IP network to the target system. When PowerHA performs a synchronization, either for the initial synchronization, after an unplanned failure, or when resuming from a suspend operation, PowerHA replicates data over the TCP/IP network.

With this enhancement, PowerHA can now compress geographic mirroring data during resynchronization reducing the data transfer required. In addition, on Power10 systems, PowerHA uses the on-chip NX GZIP accelerator if the NX feature is enabled, providing hardware accelerated compression. This enhancement will be available on June 24th, 2022 for both PowerHA 7.5 and 7.4.

Reduction in Time Required to Perform CRG Switchover for Selected Configurations 7.5 HA 5.0.1 7.4 HA 4.6.1 7.2 HA 3.8 (IBM i 7.3)

Previously, when performing a switchover of a device Cluster Resource Group (CRG), the job queue specified in the job description for the user performing the switchover was used to vary off and on the IASP. This may result in the switchover taking longer than expected. For example, if the job queue is held or only allows a specific number of active job at a time, the switchover may be held up for a long time. Previously, PowerHA included switchover progress monitoring that visually indicates a job submitted but not yet running to help identify these situations.

PowerHA has been enhanced so that planned switchovers or unplanned failovers will always use the QSYSNOMAX job queue for IASP vary on and off operations regardless of the user initiating the operation. This reduces the chance of a held job queue or hitting the restriction on the maximum number of active jobs in the subsystem, reducing the time required to perform the CRG switchover.

Recently Released Enhancements

The PowerHA product continues to provide enhancements via PTF several times a year. The following enhancements were previously released in December of 2021 and are also available in 7.5 PowerHA:

Simplified IP Address Switching 7.5 HA 5.0.1 7.4 HA 4.6.1

Previously, in a PowerHA environment, users could configure a server takeover IP address associated with an IASP device in a device cluster resource group (CRG). This IP address follows the primary node in the CRG and enables users and applications to access the correct primary system at any point in time.

Environments with many IP addresses often required a custom-coded user exit program attached to the IASP vary-on and vary-off exit points. PowerHA now allows users to add up to 256 IP addresses to a device CRG by using a new device type of *IPA. This simplifies a PowerHA implementation by enabling simple switching of IP addresses without custom coding, exit programs, or scripts.

Integrated Switchover Progress Monitoring 7.5 HA 5.0.1 7.4 HA 4.6.1

Previously, PowerHA introduced switchover progress monitoring via a new command, Display CRG Activity (DSPCRGACT), which shows the progress of a switchover. This progress monitoring has been enhanced:

Integrated into the CHGCRGPRI command to automatically display progress during a switchover.

A new action required step indicates when a job queue is held, or a subsystem is inactive.

Performance Improvements 7.5 HA 5.0.1 7.4 HA 4.6.1 7.2 HA 3.8

While PowerHA has always required minimal resources on nodes in the environment, PowerHA continues this trend by providing improved performance for many operations with the following enhancements:

Performance improvements for SVC-based storage communication, reducing the time required to perform operations.

Performance improvements related to temporary user spaces and user queues with up to a 48% decrease in CPU usage, and up to a 90% decrease in the number of user spaces created for certain PowerHA operations.

Enhancement Availability