Pre-installation setup for Unstructured Data Integration

If you want to use the Unstructured Data Integration tool in watsonx.data™ integration, ensure that the following prerequisites are met.

Prerequisites

The following services are required:
  • IBM watsonx.ai™
The following services are optional:
  • IBM watsonx.data Spark is recommended for optimized processing and performance.
  • IBM® watsonx.data intelligence is recommended for unstructured data governance capabilities and retrieval.

GPU requirements

To support entity extraction, a minimum of 2 GPUs is required. If you don't plan to use entity extraction specifically, then base extraction can run on CPUs. You might need to scale CPUs if required.