The Iceberg ecosystem today (Anders Swanson)

Описание: The data industry is moving fast toward open standards, and Apache Iceberg is at the center of it. In this episode, Tristan Handy sits down with Anders Swanson (Developer Experience Advocate at dbt Labs) to break down what’s actually working in the Iceberg ecosystem today, what’s still hard in production, and why the catalog layer has become the new battleground for cross-platform analytics.

They cover the core building blocks (query engines, object stores, catalogs), the rise of external catalogs as a fourth namespace tier, and a pragmatic phased adoption model for Iceberg integrations. Anders also demystifies vended credentials, explains why metadata performance and resiliency are non-negotiable, and digs into the still-unsolved challenges around identity and grants across platforms.

Chapters

00:00 Intro: why open standards are accelerating

01:20 What practitioners can expect from Iceberg in production

05:00 Lightning round: query engine, object store, catalog

06:20 Internal vs external catalogs

09:30 Four-part namespace and catalog-link abstractions

11:30 Metadata performance, resiliency, and caching

19:10 Iceberg integration phases

24:10 Producer vs consumer model (cross-platform mesh)

29:10 Vended credentials explained

33:30 The hard part: grants and global identity

46:50 “The making of a treaty is the treaty”

51:50 What Anders is excited for next year