pg_lake vs Trino

pg_lake: Postgres extensions from Snowflake that read and write Iceberg tables and lake files, executed via DuckDB. Trino: Fast distributed SQL query engine for federated analytics across lakes and databases.

key differences

Trino is built for larger datasets (Multi-PB) than pg_lake (≤ 10 TB).
Horizontal scaling is none in pg_lake and native in Trino.
Transactions: pg_lake offers full ACID; Trino offers limited.
Only pg_lake has transactions, columnar storage, durable persistence, compression.
Only Trino has materialized views, role-based access, time travel.

choose pg-lake for

+Giving Postgres apps access to an Iceberg lake
+Exporting Postgres data to Parquet/Iceberg
+Small lakehouse on Postgres
+Querying raw files in S3 from SQL

best at · Lakehouse, OLAP, Warehouse

choose trino for

+Open lakehouse SQL on Iceberg/Delta
+Federated analytics across databases
+Interactive BI on object storage
+Data mesh query layer

best at · OLAP, Lakehouse, Warehouse

pg_lakePostgres extensions from Snowflake that read and write Iceberg tables and lake files, executed via DuckDB.

TrinoFast distributed SQL query engine for federated analytics across lakes and databases.

overview

Primary model

Relational

Also supports

Columnar / OLAP, Spatial

Columnar / OLAP

Type

Extension

Database engine

Status

Active

Developer

Snowflake (Snowflake Labs)

Trino Software Foundation

First release

2025

2019

Latest version

3.5.9 (Oct 25, 2026)

484 (Oct 16, 2026)

License

Open source (permissive)Apache License 2.0

Free tier

✓

popularity

WhichDB Score

15.2

49.9

Rank

#220

#27

GitHub stars

1.7K

13.4K

Stack Overflow questions

–

893

scale & guarantees

Comfortable data size

Medium (≤ 10 TB)

Massive (Multi-PB)

Horizontal scaling

None

Native

Multi-region

None

ACID

Full

Limited

Consistency

Strong

Schema

Strict

Architecture

Single node, Client server

Distributed, Client server

deployment

Self-hosted

✓

Vendor-managed cloud

—

Serverless

—

Embedded

—

Edge

—

Clouds

–

capabilities

Transactions

✓

—

Joins

✓

Secondary indexes

—

Full-text search

—

Vector search

—

Geospatial

✓

JSON

✓

Change data capture

—

Triggers

—

Stored procedures

—

Materialized views

—

✓

TTL / expiry

—

Pub/Sub

—

Columnar storage

✓

—

In-memory

—

Durable persistence

✓

—

Compression

✓

—

Encryption at rest

—

Role-based access

—

✓

Time travel

—

✓

Branching

—

Online schema changes

—

Streaming ingest

—

Offline sync

—

Runs in WebAssembly

—

interfaces

Query languages

SQL, PL/pgSQL

SQL (ANSI)

Client languages

C, Java, Python, JavaScript, Go

Java, Python, Go, JavaScript, Rust, C#

Best for

Lakehouse, OLAP, Warehouse

OLAP, Lakehouse, Warehouse

Add more to this comparison Not sure either fits? Try smart match

pg_lake vs Trino: FAQ

What is the difference between pg_lake and Trino?

pg_lake: Postgres extensions from Snowflake that read and write Iceberg tables and lake files, executed via DuckDB. Trino: Fast distributed SQL query engine for federated analytics across lakes and databases. Trino is built for larger datasets (Multi-PB) than pg_lake (≤ 10 TB). Horizontal scaling is none in pg_lake and native in Trino. Transactions: pg_lake offers full ACID; Trino offers limited.

When should I choose pg_lake over Trino?

Choose pg_lake for giving Postgres apps access to an Iceberg lake, exporting Postgres data to Parquet/Iceberg and small lakehouse on Postgres. Choose Trino for open lakehouse SQL on Iceberg/Delta, federated analytics across databases and interactive BI on object storage.

Is pg_lake or Trino more popular?

Trino ranks #27 on the WhichDB Score (49.9/100) and pg_lake ranks #220 (15.2/100). Popularity is not fit: check the requirements that matter for your project.

Are pg_lake and Trino open source?

Yes. pg_lake is open source. License: Apache License 2.0. Yes. Trino is open source. License: Apache License 2.0.