What’s New With BRMS, PowerHA, And Db2 Mirror
April 28, 2021 Alex Woodie
The latest IBM i Technology Refreshes (TRs) brought a number of enhancements to various parts of the operating system and related licensed products. That includes new stuff for its backup and recovery product, BRMS, as well as the two high availability offerings, PowerHA and Db2 Mirror.
On the Backup and Recovery Media Services (BRMS) front, IBM added several new features with IBM i 7.4 TR4 and IBM i 7.3 TR10, including new SQL services that returns log information for BRMS activities. This new SQL service generates the same information that the Display Log for BRM (DSPLOGBRM) command does, but it does so through the power of SQL.
IBM i customers are starting to point their IBM i backups to the cloud in greater numbers. In particular, they’re using the Cloud Storage Solutions, which allows IBM i customers to store their backups in an object storage system in the IBM Cloud. The new TR bolsters this type of use case by improving cloud support for network installations using network file systems, IBM says.
Customers running BRMS network sync jobs will notice an improvement in performance with these releases. IBM says it goosed the network sync jobs by reducing the number of journal entries processed. Lastly, IBM delivered support for running full system backups using IBM i iSCSI supported devices.
PowerHA SystemMirror for i gets several enhancements, too, including a streamlined switchover process, courtesy of a new “progress bar” that appears on the screen. According to IBM i Chief Architect Steve Will, this is part of IBM’s focus on simplifying automation.
“PowerHA is working with some clustering technology that we also updated in the base operating system to give you a progress level of how your switchover is happening when you’re doing a move from primary to secondary to swap roles,” Will told IT Jungle earlier this month. “So there’s a monitor now that shows you the progress. Many of our clients who are using these technologies just feel better, like all of us do, if they can see in advanced that the bar” is moving toward completion.
PowerHA users will also have a better experience when utilizing FlashCopy techniques, courtesy of IBM’s FlashSystem technology. The FlashCopy commands have been integrated into the PowerHA software, eliminating the need to use external tools and scripts to accomplish the FlashCopy operation, according to IBM.
Finally, IBM says it has continued to improve the interactive prompting within PowerHA, which the company says makes it easier to manage the software. It also worked to enhance contextual F4 prompting and to optimize the suggestions for specific environments and configurations. All PowerHA enhancements will become available in June.
Db2 Mirror, IBM’s continuous availability solution that is only supported on IBM i 7.4, was also updated with enhancements in the areas of security, database performance, and object support.
Security has been enhanced with an update to the Analyze Profile Activity (ANZPRFACT) command. According to IBM, prior to this enhancement, ANZPRFACT would consider activity only on the node where the command executed.
“This could cause active profiles on the other node to become disabled,” IBM says. “This enhancement will now be multimode aware and disable profiles only if there is no activity on either Db2 Mirror node. ”
On the performance front, IBM made various tweaks under the hood to increase the Db2 for i database’s ability to replicate data in a Db2 Mirror environment. According to Scott Forstie, IBM’s business architect and Db2 for i guru, users don’t have to do anything to get faster record-level replication.
“Sometimes we’re just making internal improvements that you don’t have to do anything to adopt,” he says. “It’s the sort of thing [you get] when you hear us talk about, hey it’s in your best interest to stay current. If you just stay current with PTF group level, you’d say, ‘well, things seem faster this time.’ That’s one of these situations. ”
Db2 Mirror also now supports data queue entries, in addition to the data queue itself. It also now supports user index and user space objects. IBM decided there was enough usage of these objects in Db2 Mirror environments that it warranted building replication support for them.
Decisions like which objects to support in Db2 Mirror replication are based on user feedback, according to Forstie.
“You always have to put a little context around it, too. Do you have these objects and do they change?” he says. “If you do have them and they are not changing, then there’s probably not a lot of work for the replication engine to do. But for data que messages are being sent to them and received and extracted, so those are quite popular. Before this we only had what we called definition only. Now we have full data replication as an option, too.”
User spaces and user indexes are not used as much in the installed base, but they do exist, Forstie says. “They are less used, but they’re used enough that it was important to add them into the product set,” he says. “A lot of people use user spaces and user indexes to have a very fast lookup. Maybe the objects are not changing that frequently. But they can do memory-resident references to data, which of course you can’t do with a Db2 for i table, so [user indexes] gives them a unique performance advantage and now they’ll have the ability to have that be included in replication.”