Postgres databases grow — and most of that data sits on expensive SSD storage that nobody's queried in months. ColdFront is pgEdge's open source transparent tiering system that moves cold data to Apache Iceberg on cloud object storage (up to 90% cheaper storage) while keeping OLTP, analytics, and AI workloads running against the same tables, using the same SQL and same connection string. No application changes. No vendor lock-in.

In this live webinar, Jimmy Angelakos (Staff Software Engineer and lead developer behind ColdFront) will walk through the project itself as well as the engineering decisions behind it, and Paul Rothrock (Solutions Engineer) will deliver a live demo: from setup through creating dual-mode (hot/cold) tables, Iceberg-only tables, and traditional partition management, with reads, writes, and deletes against cold data.

Key topics:

- Why existing approaches (ETL pipelines, proprietary tiering, dump-to-files) each have a structural problem

- How ColdFront routes queries transparently across hot Postgres partitions and cold Iceberg files — without a sidecar process or remote procedure call

- The writable cold tier: why read-only archival creates real operational problems for GDPR, auditing, and corrections, and how ColdFront handles it

- Architecture walkthrough: C extension, pg_duckdb (in-process), Archiver Go binary, Lakekeeper, and why 'as few moving parts as possible' drove every choice

- Getting multi-writer correctness right: serialising concurrent cold writes across pgEdge Spock nodes with a Lamport-style bakery protocol, model-checked in TLA+

- Live demo: setup, dual-mode tables, Iceberg-only tables, traditional partitions, archiving, and querying across both tiers

Give ColdFront a try anytime: https://github.com/pgEdge/ColdFront

Fill out the form to watch this session on-demandRegister below to reserve your spot today!