Overview of Data Gate for watsonx

Edit online

Data Gate for watsonx is the most efficient way to bring transactional data from an IBM Z® mainframe computer to IBM watsonx.data.

With this valuable transactional data, organizations can better identify fraud, understand constituent behavior and customer buying journeys including customer attrition, and build predictive AI models to understand, anticipate, and influence business outcomes. By bringing transactional data that originates on the mainframe into an open, governed data lakehouse like watsonx.data, an organization can build AI models to drive greater revenue, improve productivity, and better manage cost.

Transactional data plays a crucial role in the era of cloud computing and artificial intelligence (AI):
  • Transactional data is essential to understand the current state of the business.
  • Transactional data is the most valuable data for the prediction of business outcomes.
  • Most enterprise organizations keep their transactional data on IBM Z.

Data Gate for watsonx gives analytics and AI applications access to transactional data on IBM Z by making this data available to watsonx.data.

watsonx.data can provide large Iceberg-type data lakehouses that can serve as the basis for extensive analytics and AI computations. In addition, watsonx.data can offer an SQL view on nearly all types of data, structured and unstructured.

IBM Data Gate for watsonx infuses watsonx.data with data from Db2® for z/OS®, VSAM, or IMS data sources, so that you can use the capabilities of watsonx.data to extract insights from transactional data on a mainframe.

The IBM Z data is not moved. Data Gate for watsonx copies it 1:1 to watsonx.data and synchronizes (replicates) the copied data continuously, which can ensure consistent query results on both sides.

The copying of the data is a unidirectional process. The data flow is always from IBM Z to watsonx.data, but not the other way around. This way, applications at the other end cannot overwrite or compromise data on the IBM Z, keeping it safe. The operational impact on IBM Z transactional processing is low; operations can continue as they used to.

The copying process is highly parallelized to allow for maximum performance and throughput. Code page conversions are handled automatically and efficiently in the background.

Benefits of Data Gate for watsonx

Data Gate for watsonx delivers comprehensive value to IBM Z customers through multiple key advantages:

  • Modernizes mainframe access for analytics and AI applications by making Z data readily available to watsonx.data
  • Delivers Z customers a streamlined, efficient journey to watsonx.data with the minimal resource consumption, latency, and highest possible throughput compared to other replication options
  • Delivers first-class integration of IBM Z data in watsonx.data
  • Outperforms competitive offerings with tangible added value for Z data
  • Enables high-performance replication with IBM Integrated Synchronization or IBM InfoSphere® Classic CDC for z/OS (CDC)
  • Maintains low footprint on Z and ZIIP eligibility
  • Enables transactionally consistent queries in a synchronized cloud
  • Supports containerized cloud-based application
Figure 1. Components and data flow
Diagram: components and data flow

Example workflow for provisioning a Data Gate for watsonx instance

The figure shows the wizard for provisioning a Data Gate for watsonx instance. The Data Gate for watsonx user interface (UI) that includes this wizard runs on Cloud Pak for Data.

Figure 2. Provisioning a Data Gate for watsonx instance from the UI
A screen capture of the UI wizard for provisioning Data Gate for watsonx instance

These steps are needed to complete on the UI:

Select the data source
The supported data sources are:
  • Db2 for z/OS
  • IMS (Information Management System)
  • VSAM (IBM Virtual Storage Access Method)
Select the target
An object stored (bucket) in watsonx.data.

Create a target instance if it does not exist on the cluster. Your Cloud Pak for Data user ID is used to access the target data, which requires the admin role.

Provide access
Provide the access key and the secret key for accessing the object store.
Validate credentials
Validate the watsonx.data target information and login credentials.
Define resources
If your data source is Db2 for z/OS, define Data Gate for watsonx resources (CPU, memory, and namespace).
Confirm
Confirm your choice and start provisioning.

For a detailed description, see Provisioning Data Gate for watsonx.