Agentic AI Reference Architecture

Orchestration · Data artifact

Extraction Cadence Policy

Data artifactOrchestrationOrchestration & ToolsVariation point (abstract)arc:ExtractionCadencePolicy

An abstract configuration choosing whether a pipeline extracts source data in periodic batches or continuously as it arrives.

Responsibility. Fixes the extraction cadence and hence knowledge freshness of a pipeline.

Also known as: Batch vs streaming extraction decision

configuresis specialized byis specialized byIngestion Pipeline Orchestrator: configuresIngestion Pipeline Orche…Batch Extraction Policy: is specialized byBatch Extraction PolicyStreaming Extraction Policy: is specialized byStreaming Extraction Pol…
Direct neighbourhood (hover for relationship types)

Variants

VariantWhen to choose
Batch Extraction PolicyChoose for relatively static knowledge bases where daily updates suffice, and for structured sources with efficient timestamp-based change detection.
Streaming Extraction PolicyChoose for rapidly changing domains (e.g., news, stock trading) and semi-structured event sources where near-real-time freshness justifies added complexity.

Relationships

configures structural

Sources

  1. Ch6.3A: T. Nguyen, "ETL Pipeline Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 6.3A. ISBN: 9798244538229.