Data lakehouse

/02 About

Batch ingestion into raw and curated zones, governed transformation, a serving layer for analytics and published data products.

Operational and external data lands unchanged in a raw zone, is cleaned and conformed into a curated zone by orchestrated transformations, and is served to analysts and applications through a warehouse-style serving layer and named data products. When to use: analytics and reporting across many sources, machine-learning feature preparation, and organizations that want one governed copy of analytical data. Trade-offs: cheap storage and many workloads on one copy, with strong lineage; it needs data ownership, quality checks and a catalog to stay trustworthy. Latency is minutes to hours; use the streaming pattern when you need seconds.

Published by Lattix · 12 elements · 12 relationships · validated on publish