Rocket CDC Tool Pushes Data Out Of IBM i

March 10, 2025 Alex Woodie

IBM i shops that need a change data capture (CDC) tool to push Db2 data off the box onto a target system, such as an Apache Kafka stream, may want to check out the new DataEdge product that Rocket Software recently announced.

In late January, Rocket Software unveiled DataEdge, a new suite that provides a spectrum of data integration, data discovery, and data virtualization capabilities under the specific product names Rocket Data Replicate & Sync, Rocket Data Intelligence, and Rocket Data Virtualization. Some of the components in DataEdge are preexisting, while others – such as the outbound IBM i CDC capability in Data Replicate & Sync – are brand new.

Rocket’s goal with bringing the three products together in the DataEdge suite is to help customers modernize their data environments in preparation for higher order projects, including advanced analytics and AI.

Organizations may not always have a good grasp of exactly how data flows among their various systems, which may include sources like IBM hosts and Windows and Linux-systems, and various data warehouses running on prem or in the cloud. With DataEdge, customers can understand what data flows they have, create new mappings to support new data integrations, and then execute those integrations, in some cases using a data virtualization approach.

DataEdge is all about getting data where it can have the most impact, says Don Spoerke, a product manager with Rocket.

“Real-time data access is really going to be one of the most important commodities of any organization,” Spoerke says. “The data lake or data warehouse, the latency – even though it may even sometimes be down to an hour or so –just may not be good enough.”

The Rocket Data Replicate & Sync product provides data real-time connectivity for a range of platforms, including IBM hosts. On the IBM i platform, it has supported inbound CDC connections – where external data is brought to the Db2 for i database – for about fifteen months, Spoerke says. “Having IBM i as a source was just announced and released back in December,” he tells IT Jungle.

The benefit of outbound CDC capabilities on IBM i and System z hosts is that it doesn’t consume CPWs or MIPS on the IBM hosts, he says. All of the data processing for pulling the real time data off the IBM hosts is born by the external system, typically a Windows or Linux environment. Getting that valuable transactional data off the host and onto another system can provide a number of benefits for customers, he says.

“I can now take Db2 data and put it out into Kafka, for example. And then once that out, once it’s out in Kafka, I can do whatever I want to it,” he says. “Once I can get my data synchronized to almost zero latency, now that data can be consumed by whatever application I want without having an impact on my production system, because last thing you want to do is run a bunch of additional queries against an iSeries or even a zSeries box, because you’re sending big checks to IBM.”

DataEdge also brings IBM i and System z shops benefits in the field of data discovery. Rocket Data Intelligence can interrogate host programs to determine how the programs work, what files they’re touching, and how changes in one part of the program can impact others, says Dani Liu, the product manager director with Rocket.

“One the discovery side, we’re driven by the metadata discovery platform, so we cover the deepest and broadest sources, including IBM mainframe cloud – whatever customers have,” Liu tells IT Jungle.

“Specifically in terms of IBM i, we understand what they have, what is referencing what, what programs can be touched, what programs can’t, and what data is redundant and what processes are important,” she continues. “We can understand SQL scripting language and COBOL and RPG code running on the system, and we can inspect some of the physical file definitions to make sure that everything is traced and tracked, so that way customers have confidence and there’s less risk in terms of modernization.”

Rocket has tried to help some customers modernize their data environments, but they’re having a real tough time, Liu says. “They’re just at loss in terms of they don’t know what they have,” Liu says. “They don’t know what’s connected.”

The automated data discovery capabilities of the Data Intelligence product, which Rocket launched in June 2023, can provide a place to start, whether it’s a file, a program, or a specific job, she says. Once the software has automatically scanned the system and provided its report, the customers can decide how they want to move forward. That may include using the Rocket Data Replicate & Sync software to create new data integration pipelines to Snowflake or Databricks lakehouses, for example, or it could mean using Rocket Data Virtualization.

Data virtualization is a data management technique that essentially creates an abstraction layer atop existing data sources. It doesn’t really apply much in the IBM i environment, Spoerke says, but it does apply in the IBM mainframe world, where non-relational data types, such as VSAM, IMS, DATACOM/DB, and Adabas data sources need to be tapped into.

“Those are not traditionally SQL accessible,” he says. “So by building a virtual descriptor on top of that, we make it look like a relational table, because we’re going to take that take that inbound SQL, we’re going to manipulate it and turn it into the native calls for the databases.”

Customers can pick and choose which elements of DataEdge they want to adopt. The company isn’t trying to solve all of their customers’ data problems with the tool, but they did see enough commonality with their customers and what they’re struggling with to feel confident that the tool will resonate. Beyond data discovery, data integration, and data virtualization, Rocket will partner with other customers.

“We want to adapt to where customers are at and work with our ecosystem, so we actually partner with other data catalogs and quality tools,” Liu says. “They can actually take the lineage information, especially on the IBM i or mainframe side, and also import whatever governance and context they have into the tool of their choice, so that’s the flexibility that we’re offering with Data Edge.”

This Issue Sponsored By

Table of Contents

Content archive

Recent Posts

Subscribe

Pages

Search