SYSTEM Cited by 1 source
Infor Data Fabric Stream Pipelines¶
Infor Data Fabric Stream Pipelines is an add-on real-time streaming feature inside Infor Cloud ERP that captures change events from configured ERP tables (sales orders, inventory, financial transactions, etc.) and publishes them immediately — "without waiting for storage in the data lake" — to external subscribers over HTTPS in NDJSON.
Mechanism (as disclosed in sources)¶
From Oldcastle's production deployment:
- Enabled as an add-on feature within the Infor Cloud ERP environment.
- Operator configures which ERP tables stream (per-table opt-in).
- Captures insert, update, and delete operations with metadata (operation type + timestamp). This is a standard CDC shape.
- Emits change events "immediately upon data modification" — no periodic batch window.
Seen in¶
- sources/2026-04-21-aws-oldcastle-infor-aurora-quicksight-real-time-analytics — Oldcastle APG uses Stream Pipelines as the source tier feeding an AWS streaming analytics stack: events flow via HTTPS (port 443) into a static-Elastic-IP NLB, through an EC2 NAT router, past RDS Proxy, into Amazon Aurora PostgreSQL JSONB columns (see patterns/nat-router-for-static-ip-ingress + patterns/streaming-cdc-to-relational-buffer).
What the source does not disclose¶
- Delivery semantics — at-least-once? exactly-once? Not specified.
- Backpressure — behavior when downstream HTTPS endpoint is slow or returning errors not specified.
- Schema-change signalling — whether schema evolution on the source ERP table surfaces in the stream's metadata is not discussed.
- Throughput ceilings — no rate limits or event-rate numbers disclosed.