Announcing IBM i 7.5, 7.4 TR6, IBM i Merlin

IBM i 7.5: Part 1

I’m going to tell the IBM i 7.5 story in two parts. But before I even get to the first part, let me remind you that there are details about everything in the release on the “support pages.”

Db2 Enhancements

Db2 for i is the core of our platform—always has been, always will be. In the old days, every major release had some of its biggest buzz from the big new features from Db2. These days, Db2 buzz is almost constant, though—because like the rest of IBM i, Db2 puts out new features twice a year, and sometimes more often than that. So, before I get to the 7.5 features in Db2, let me also tell you that the team has put out new functions in 7.4 for database engineers (DBEs) and SQL programmers, including improved generation of SQL DDL, an improved ability to use CL with SQL scripts, and better usability and new capabilities in SQL HTTP functions. These 7.4 enhancements, like the base 7.5 enhancements, have their own starting page.

Now, back to 7.5 Db2:

• BOOLEAN: Db2 for i now supports the SQL standard for BOOLEAN as a data type in 7.5. And, as you’d expect, RPG and JSON both can take advantage of it immediately.
• Huge jump in the maximum size of a Binary Radix index—to 16 TB! Perhaps you were fine with the previous limit (1.7 TB) but we have many customers who use IBM i for workloads where the amount of data is growing immense. We’re staying ahead of their needs with this increase.
• Ragged flash support—if you’ve been a Flashcopy user, you’re familiar with the fact that you’ve always had your pending transactions rolled back on the server, which is the target of the flash. In other words, if you were making changes to your data—if transactions were going on in Db2 when the Flashcopy happened—you’ve never been able to ensure that the target data is exactly the same as the source.  With 7.5, Db2 allows you to tell the Flashcopy to NOT roll back the pending transactions. This provides you with an opportunity to start from the exact same point as when the Flashcopy occurred. This has been a big request from customers, and now that option is available.

Db2 Mirror Enhancements

The IBM Db2 Mirror for i product was initially introduced with the IBM i 7.4 release. The clients who’ve adopted it so far are benefitting from the continuous availability it provides their application environments. In 7.5, there are three big features which allow Db2 Mirror to fully realize its potential:

• Mixed Release Support: The two IBM i instances in the mirrored pair can be at different major release levels. This is a key enabler to “rolling upgrades,” where you can keep your applications running on one 7.4 system while you upgrade the other system in the pair to 7.5. When you bring the new 7.5 instance up, it can re-join the mirrored pair, move all of the work to the 7.5 system, upgrade the second system, resynchronize and you’re done. This means zero downtime moving from one release level to another. You can even leave one system at 7.4 for a while, and move when it’s convenient. Of course, there may be some things in 7.5 you can’t do while it’s paired to a 7.4 system, so we would not recommend leaving them at different releases for a long time, but for many clients, a few hours or days is enough.
• Active/Read-Only Mode: Many clients want to use Db2 Mirror to have one system be the “active” system—the production system where the transactions take place—while the other is ready for immediate takeover in an emergency, but this second system is locked down so that it cannot make changes which will get mirrored to the production system. As of 7.5, this type of behavior can be enforced by Db2 Mirror.
• Db2 Mirror Versioning Service: Internally, we’ve created a means to track differences in code levels and functions—it’s how we are able to do “mixed release” above. Well, Db2 Mirror has now externalized this function, so applications can also track differences in their own code levels and functions. This feature is also being made available in the 7.4 version of Db2 Mirror.

Security Enhancements

The number of changes we’ve made to allow easier security in 7.5 is astonishing. In general, they fall into two categories, and I can only shine a light through the doorway into a huge room you’ll be able to explore.

• Passwords: You can now encrypt passwords with a much stronger encryption scheme (SHA2-512), you can use an API to check whether a potential password will meet all the password rules you have in place, without changing the password, and you can be assured that “bad actors” will not be able to tell if they got the userid wrong, or the password wrong when an authentication fails.
• Default *PUBLIC Authority values changed to *USE: Throughout the system, we’ve found many objects which were shipped with *PUBLIC being *CHANGE. And, while that didn’t really open holes in the integrity of the system, it was hard to explain that to auditors. So, we’ve changed it to *USE in seemingly countless places. Again, this is highly unlikely to make things behave differently, but it should make audits much simpler.

IBM i Merlin

We’ve been working on a product to help our clients move into the modern, cloud-capable world, and in addition to IBM i 7.5, today is the announcement day for the product with the official name: IBM i Modernization Engine for Lifecycle Integration.

It’s a long name, like so many IBM product names, but internally we’ve been calling it “Merlin” and that’s flowed out into the product in several ways. I will probably write about those in a future blog post, but for now, let me tell you a bit about what it is.

Merlin is a “wizard of wizards”—and most of us know what wizards are—in the IT world, they are tools (applications) which help you do something, and they make it simple.

Merlin, then, is a set of tools. The first kinds of tools we’re introducing with Merlin are tools that help you develop software with the “IBM i Next Gen Apps” approach (remember that blog post?)

This means that the first release of Merlin focuses on developing code, so it has an integrated development environment (IDE), using a DevOps approach (so it has wizards to help you set up a DevOps environment, or to integrate your IBM i development into an existing DevOps environment), while making it easy to publish and consume function using services. And it does all of this with a browser interface.

Yes, rather than installing these tools on a PC for each developer, these tools run in OpenShift containers, and once they are set up, users (typically programmers for this initial set of tools) get at them with a browser.

I will have to go into further details about IBM i Merlin in future posts, but maybe now is a good time to tell you that there are webcasts this week which cover

• An overview of all the announcements
• An overview of IBM i Merlin
• The latest on security and Db2 for i

All of these webcasts can be found here.

The features of this first release of Merlin are just the start of a much longer roadmap of tools we want to provide which will help customers simplify their modern use of IBM i. Take a look at what’s there and then start providing ideas about what should come next (and you don’t have to bother saying “a debugger.” We know. It’s coming …)

IBM i 7.5: Part 2

Oh my goodness, look at the time! Or rather, the word count! As I expected, I can’t write a short post when we have such a big set of announcements.  I’ll highlight the highlights of other big things:

• We’re using the on-chip “Nest Accelerator” (aka “NX”) on the Power10 systems to implement ZLIB compression in several ways, but perhaps most notably to answer the highly requested RFE which asked us to “add a modern compression algorithm for SAV/RST.”
• Full support for up to 48 cores in a partition of IBM i 7.5—with SMT8 that’s up to 384 threads! And if you need something bigger than that (and we have some customers who do) our Lab Services team can help you get to 240 processors and 1920 threads in a partition.
• The job scheduler, which is part of IBM i supports *YEARLY
• Access Client Solutions (ACS) has a bunch of new stuff, making it easier to work with multiple things at once: SQL Script tabs, systems in defined groups and more.
• New Nav, new stuff! And of course, though the new IBM i Navigator can work with any supported release, there are big new functions available concurrent with the 7.5 announcement including Performance Data Investigator! Now that “New Nav” is THE Nav—it’s the default and only Navigator—you’ll want to take advantage of the power and simplicity it provides.

OK, that’s enough for me now.  Until next time—be well, and do great things!