Data Henrik on Data Henrik - Life in IT

How to search IBM Documentation

Tue, 16 Jun 2026 14:34:00 +0200

Back in 2021, in my blog post My best practices: How to search Db2 documentation, started like this:

I often answer technical product questions, for Db2 and IBM Cloud topics. To provide relevant links and to back up my “hunch” after reading a question, I typically search the relevant documentation. But what are efficient ways to search in the Db2 documentation? What are good ways to find the relevant parts in the IBM Cloud documentation? Here are my best practices on searching documentation.

IBM Bob explains a Db2 vector query

Thu, 21 May 2026 15:47:00 +0200

In my previous blog post on Db2 vector index support, I included two queries with their respective access plans. For the sake of testing my options, I asked IBM Bob to explain to me in simple English the “explained query” that I provide to you as r24_nested.sql.explained.

Here is my input as prompt:

explain, in simple English, the Db2 access plan and “explain output” available in file @r24_nested.sql.explained

And here, for your pleasure / education / interest, the answer I received.

Soon as your neighbors: Db2 Vector Indexes

Tue, 19 May 2026 17:12:00 +0200

In my recent blog post about interacting with LLMs from SQL with external models in Db2, I mentioned the Db2 Early Access Program (EAP). The EAP provides a preview on upcoming Db2 features. Db2 added vector support in version 12.1.2. What is still missing, is the ability to index vector data. That’s something in the development pipeline and made available in recent Db2 EAP releases. In today’s blog post, I am going to give you an introduction to vector indexes in Db2, share my test setup, and some (early) observations.

Db2: SQL-embedded LLM-based data generation (external models)

Fri, 08 May 2026 16:27:00 +0200

In my last blog post on Financial research using Db2 vector capabilities, I discussed how to utilize the vector capabilities in Db2 for looking for specific exchange-traded funds (ETFs). Similar to earlier examples, I encoded the vector embeddings with ollama by calling a function from its Python SDK. It worked ok in my scenario, a Jupyter notebook with Python and SQL code for Db2. But vector encodings were all computed outside of Db2 and the SQL context. So how would I write more complex queries?

Configure Db2 Genius Hub for repository-based authentication

Fri, 17 Apr 2026 16:57:00 +0200

Have you already worked with or even installed IBM Db2 Genius Hub? Db2 Genius Hub is a web-based tool for (not only) Db2 users that helps you understand, govern, and optimize your data within IBM Db2. One of the first steps after installation is completed should be to configure user access and privileges. Here is my write-up from a recent workshop which I helped to prepare.

Db2 Genius Hub Authentication

When installed and initially configured, Db2 Genius Hub comes with a default administrator account (admin user). That admin account is only meant for the initial setup and as a means to configure the actual authentication. Db2 Genius Hub, right now, supports three different authentication providers:

Financial research using Db2 vector capabilities

Fri, 27 Mar 2026 15:05:00 +0100

Last year, I introduced you to the new vector feature in Db2. First, I showed you how to perform similarity search, then explained how to use vector functionality for geo(-spatial) queries. Now, I switched example for similarity search to an interesting use case, performing research on ETFs (Exchange-traded funds), or mutual funds in general.

Exchange-traded funds

Many people use ETFs for retirement savings, often without looking into details. An ETF can have many properties, including the market sectors, geographic focus, asset and industry types, risk level, and many more. So, knowing one specific ETF, how do you find similar ETFs? Well, use similarity search…

Pi Day with Db2

Mon, 09 Mar 2026 16:45:00 +0100

Pi Day

Many things are small, but important. The mathematical constant π (Pi) is one of those cornerstones. It is used in many places in mathematics, physics, and more. The constant has never ending digits (“infinite”). Some kids, mine included, have fun memorizing 100, sometimes 200, or even more of those digits by singing along songs like the following:

To celebrate π, fans started to observe a Pi Day on March 14th (3.14). According to Wikipedia, the first larger Pi Day was celebrated in 1988. People eat pies, recite π and its digits, or have other fun with π.

Db2 12.1.4 and Db2 Genius Hub are available

Fri, 06 Mar 2026 11:13:00 +0100

Db2 12.1.4 for Linux, UNIX, and Windows

Yesterday, the latest update to Db2 Version 12.1 became available. You can download it from the page Download Db2 fix packs by version for Db2 for Linux, UNIX and Windows. The Db2 documentation has a What’s new in Db2 12.1.4 with the usual highlights and subpages with more details. Part of the news is the availabilty of Datalake tables outside containerized deployments.

As usual, I checked that the Db2 Community Edition for Docker has been updated and it was:

Getting started: IBM Bob and Db2 Developer Extension combined

Thu, 05 Mar 2026 14:24:00 +0100

Getting Started

Recently, IBM released the IBM Db2 Developer Extension for Visual Studio Code. It allows you to easily install a Db2 instance based on the Db2 Community Edition, catalog Db2 connections, and execute SQL statement. This could be done without the extension. The real value is syntax checking and highlighting, code completion while editing SQL statement, and more.

You can install the extension from the marketplace (see link above) or by downloading the VSIX file from the Db2 Developer Extension’s GitHub release page. Once downloaded, run the command “Extensions: Install from VSIX” and pick the VSIX file. The latter is also the way I installed the extension into IBM Bob. Bob is currently in tech preview and its product page states: “Bob is your AI software development partner that understands your intent, repo, and security standards.” Bob offers some predefined modes and personas and acts as a AI-powered partner for the entire software development lifecycle. As such, you can plan a project, code, maintain software, research documentation or features, and more. Bob is based on VS Code and, hence, you can utilize most extensions.

Db2 Valentine's Day Special

Wed, 11 Feb 2026 14:42:00 +0100

Valentine’s Day

Valentine’s Day is coming up and it is time to get prepared. In the past, I had written about how to impress your spouse and how to date with Db2. Today, I want to expand on this, really. I am going to show you how to teach Db2 to expand bits and bytes into ASCII art. Some input data and some simple SELECT statement and, voila, Db2 shows a heart and its love to whoever is seeing it. So let’s get started…

Db2 and UUID explained

Thu, 22 Jan 2026 16:23:00 +0100

New UUID functions in Db2 12.1.3

When I looked over the section “What’s new in Db2 12.1.3”, I saw that two functions to generate UUIDs (universally unique identifiers) are listed as part of the SQL enhancements. The functions are GENERATE_UUID and GENERATE_UUID_BINARY.

The documentation for the first function states:

The GENERATE_UUID function returns the formatted string representation of a Universally Unique Identifier (UUID) using the version 4 algorithm.

What exactly is a UUID and what does “version 4 algrothm” mean? I will discuss this later. First, let’s take a look at the function and its output:

Db2 and JSON array access

Mon, 15 Dec 2025 09:11:00 +0100

Db2 JSON Support

Since many years, starting with Db2 11.1.4.4 has improved support for querying JSON data. At that time, I wrote in the linked blog post:

Application developers benefit from new JSON support that features functions like JSON_QUERY, JSON_TABLE and JSON_EXISTS (Hello, XML developers!).

At the core, there is JSON_QUERY to return a SQL/JSON value from the provided JSON value based on a SQL/JSON path expression. The function JSON_TABLE does the same, but returns a table. It allows to specify a path to the root of relevant data within the queries JSON objects, then a set of paths to extract the values making up the resulting columns. I gave a talk on that topic and used this Jupyter Notebook to demonstrate the capabilities.

Past IDUG EMEA conferences

Thu, 20 Nov 2025 14:23:00 +0100

IDUG EMEA conferences of the past

IDUG, the International Db2 User Group, or better the community of customers, partners and consultants, and IBMers celebrated its 35th anniversary at the IDUG EMEA 2023 conference in Prague. The birthday cake also had logos for 40 years of Db2 for z/OS and 30 years for Db2 LUW (Linux, UNIX, and Windows).

But where have been those conferences held? Here is my attempt to document the EMEA conferences. It is based on this slide shown at the IDUG EMEA 2018 conference in St. Julians, Malta, and my own material since then.

Solving Linux suspend not working

Tue, 11 Nov 2025 18:32:00 +0100

Linux System Suspend

Usually, at the end of a workday or when moving between places, I suspend my Linux-based laptop. I simply press the sleep button on my external keyboard or click on the suspend button in Gnome. Today, however, this did not work. My machine was blocked from going to suspend. Was there any other option to suspend it? Or could I find out what as blocking it?

In the bash shell, I tried the command line option (which I found): systemctl suspend was supposed to suspend my Fedora system. Would it?

Db2 vector for GEO search

Fri, 07 Nov 2025 17:04:00 +0100

In my blog post Db2 vector for similarity search, I discussed a demo using embeddings for similarity search. It is a “typical” AI use case for vector capabilities. Today, I am going to show you a use case totally without artificial intelligence - no GPUs, no models, etc. required. And it is even very easy to set up.

Vectors for geographical coordinates

Here on earth, every point on the surface can be located by its latitude and longitude (more or less). The pair makes up the geographical coordinates of a point on a map. And, in practice for our demo, is a nice two-dimensional vector. Thus, it should be an easy to set up scenario to demonstrate the Db2 vector data type and its distance search.

Db2 V12.1 Mod Pack3 (12.1.3) is available

Thu, 06 Nov 2025 15:41:00 +0100

Db2 Version 12.1 Mod 3 Fix Pack 0 for Linux, UNIX, and Windows

Yesterday, the latest update to Db2 Version 12.1 became available. You can download it from the page Download Db2 fix packs by version for Db2 for Linux, UNIX and Windows. The Db2 documentation has a What’s new in Db2 12.1.3 with highlights. Moreover, the subpages detail enhancements related to performance, security, manageability, and more.

Among the mentioned highlights are the enhanced AI Query Optimizer, more functionality around the vector data type (see Db2 vector for similarity search), functions for generation of UUIDs, and usability / manageability improvements.

Next DeDUG meeting at IBM Munich on 14.11.

Fri, 31 Oct 2025 14:22:00 +0100

DeDUG: German Db2 User Group

Are you already registered for the next DeDUG (Deutsche Db2 User Group) meeting? It’s on Friday, 14.11.2025, at the IBM Watson Tower in Munich or remote. As usual, I highly recommend attending in-person to make the most out of it and to network.

Check out the agenda available off the DeDUG website at IDUG (free IDUG registration needed) for details. Talks will cover Db2 with PaceMaker on AWS, Guardium, Db2 Vector Support, Rust programming, and more.

Db2 External Tables at IDUG EMEA 2025

Thu, 30 Oct 2025 07:05:00 +0100

Db2 External Tables

In the past, I blogged several times about the CREATE EXTERNAL TABLE and how to use Db2 external table feature. My most recent blog posts on that topic were Loading and unloading of data on remote clients with external tables and Once again: Db2 External Tables where I looked into named versa transient external tables. You will find more links in the mentioned blog posts.

Yesterday, I presented at IDUG EMEA 2025 to an interested audience with great feedback. There was one question that I answered with a “YES, absolutely!”, but realized that my slides and blog posts were not addressing it yet. The question was whether it is possible to combine data from regular Db2 tables with data from external tables. And, yes, absolutely, that is the big benefit and allows bringing in external data for/in many use cases.

Setup for Db2 vector for similarity search

Mon, 15 Sep 2025 13:04:00 +0200

In my recent blog post Db2 vector for similarity search, I discussed a demo that I built, Jupyter Notebook to showcase Db2 VECTOR data type and similarity search. In that blog post I promised to follow up with a write-up on the setup. Here we go…

Components for Db2 vector tests

For my tests of the Db2 vector data type and its vector-related functionality, I used the following components:

IBM Db2 as relational database with integrated vector capabilities
ollama with IBM Granite embedding model
Jupyter Notebook, including JupySQL magic for Jupyter Notebooks to easily run SQL statements in notebooks

The notebook acts as database application and runs SQL statements against my Db2 test database. The app utilizes an ollama service to turn product features (selected parts of a data record) into a vector, a so-called embedding.

Db2 vector for similarity search

Fri, 05 Sep 2025 14:49:00 +0200

When I wrote about the availability of Db2 V12.1 Mod Pack2 (12.1.2), I stated the following:

To me, the most important new feature is the addition of a VECTOR type and related functionality. It allows to create, store, and query vector data close to the business data. Thereby, Db2 can be used for new AI-related use cases, but also supports the classic “what is related” queries.

Today, I am going to look deeper at the VECTOR type and share with you a Jupyter notebook I created based on my tests.

Db2 security: Modes, settings, links

Wed, 20 Aug 2025 15:39:00 +0200

This blog post is one of those bookmark articles for my own memory. I hope you enjoy it or find it useful nonetheless…

Security modes in Db2

I stumbled over a new page in the Db2 12.1 documentation, titled Security modes in Db2. It shows the three levels of security which Db2 provides to help comply with the government standards such as NIST SP800-131A and FIPS 140-2. The levels are NOFIPS, FIPS compatibility, and STRICT_FIPS mode. The levels can be activated by setting the Db2 registry variable DB2AUTH, or just an environment variable on client side, or similar variables for LDAP security.

Revisited: Containerized Db2 client and connect with API key

Wed, 30 Jul 2025 18:12:00 +0200

Use an API key to connect to Db2

Some five years back, I blogged about how to use an API KEY or ACCESS TOKEN to connect to Db2 on IBM Cloud. What I had forgotten since then was the fact that I actually had utilized the Db2 Warehouse client container for that scenario. After realizing it, I set out to test the scenario again of using an IBM Cloud API key to connect from the client container to Db2 on Cloud.

Containerized Db2 client, tools, database size

Wed, 23 Jul 2025 13:17:00 +0200

Db2 Warehouse Client Container - Again

In my last blog post, I explained how to get, set up, and run the Db2 Warehouse client container and connect to a database. While working with that Db2 environment, I also read some Db2 Warehouse client container documentation. The following page attracted my interest: commands command. A command for commands?!

So I had to try it. The output starts with this text, followed by a list of commands with their short descriptions.

Easy setup: Containerized Db2 client

Fri, 18 Jul 2025 14:29:00 +0200

A follow-up to Db2 Community Edition

Around the recent German Db2 User Group (DeDUG) meeting, I had some discussions on how to easily test Db2 and options for getting the Db2 Command Line Processor (CLP) up and running. Last year, I blogged about three plus one simple and free ways to test Db2. Later, I tested the Db2 Community Edition for Docker. But, if you only need the local Db2 client because you connect to a cloud database, what would be your options?

Cheatsheet to the IBM watsonx portfolio

Mon, 30 Jun 2025 17:32:00 +0200

From IBM Watson to watsonx: History

Last week, I ran into the “We are interested in Watson, but exactly is it?” situation. Someone at a customer knew that “Watson” was the “IBM AI thing”, but wanted to be sure and asked for a quick overview. Because I use this blog for my notes, why not write down this quick cheatsheet for my own assistance (pun intended)…

IBM recently celebrated 114 years of corporate history. It started research in what today is called Artificial Intelligence (AI) in the 1950s (see 70 years advancing AI). Considering the Timeline of Artificial Intelligence, it is only after some thousand years of the initial concepts, but still early in today’s “era of AI”. And it included some well-known AI showcases like the chess computer Deep Blue in the 1990s, IBM Watson playing Jeopardy in 2011, or Project Debater in 2018/2019.

Hyperlegible fonts, more legible posts

Mon, 09 Jun 2025 15:03:00 +0200

Over the long weekend, I had some time to improve this blog. Because the blog serves as my written memory, I need to quickly find my notes and be able to read them. Thus, you may have noticed that I manually improved some of the converted blog posts and fixed the code formatting. Some old articles had HTML markup applied to individual characters and lines which screwed up the automatic conversion of my blog. I also fixed up some blog posts with embedded photos or videos, like Impressions from Zeppelin flight or 50 Years of IBM Mainframe: The Art of Selling. Not all is fixed for sure. If you find an article that needs to be improved, let me know.

Db2 V12.1 Mod Pack2 (12.1.2) is available

Fri, 06 Jun 2025 11:21:00 +0200

Db2 Version 12.1 Mod 2 Fix Pack 0 for Linux, UNIX, and Windows

Yesterday, the latest update to Db2 Version 12.1 became available. You can download it from the page Download Db2 fix packs by version for Db2 for Linux, UNIX and Windows. The Db2 documentation has a What’s new in Db2 12.1.2 with highlights and the usual subpages with some more details related to performance, security, manageability, and other enhancements.

Next DeDUG meeting at IBM Böblingen on 11.07.

Thu, 05 Jun 2025 10:39:00 +0200

DeDUG: German Db2 User Group

Are you already registered for the next DeDUG (Deutsche Db2 User Group) meeting? It’s on Friday, 11.07.2025, at the IBM Germany Lab in Böblingen or remote. I highly recommend attending in-person in Böblingen because very likely it will be the last time at the (historic) IBM campus, IBM is moving later this year.

Check out the agenda (free IDUG registration needed) for details. Talks will cover Workload Management, Db2 clients and drivers, “explain”, monitoring, and more.

Db2 system information via SQL

Tue, 27 May 2025 11:20:00 +0200

What is your Db2 build level?

Today, during a discussion, there was the question on how to find out which Db2 version is used for the Db2 on Cloud instance someone was using. There are some ways to get to that information:

Ask the IBM support.
Look in the related documentation.
Use SQL.

Opening a support ticket (if you are entitled) is probably not the best way to find the Db2 build level. So what is the Db2 product documentation you need? I would recommend the page What’s New in IBM Db2 Warehouse, IBM Db2 Warehouse on Cloud, and IBM Db2 on Cloud with links pointing to:

News from IDUG: EMEA conference registration and important podcast episode

Fri, 16 May 2025 13:41:00 +0200

IDUG EMEA conference registration

IDUG, the Internation Db2 User Group, just announced that registration for the 2025 EMEA Db2 Tech Conference (EMEA25) is now open. You can benefit from using the early bird rates from now to July 25th.

The conference is going to be at the Hilton in Düsseldorf, Germany, from Sunday, October 26, to Thursday, October 30, 2025. Regular sessions will be from Monday to Thursday with workshops and labs on the Sunday and Thursday before and after the conference.