Glossary

AI Inference Partition

A separate supported operating-system partition used to execute an AI model while an IBM i partition may remain the production application and data source.

IBM's July 2026 Power S1112 announcement included a customer example of Linux partitions running alongside IBM i to explore AI. A partition boundary lets the inference runtime be sized and governed separately from the production IBM i workload. It does not by itself establish model compatibility, accelerator support, or secure access to Db2 for i data; those need a workload-specific pilot.

Related Terms