Amazon Aurora PostgreSQL Embeds DuckDB to Query Iceberg and Parquet Data in S3, the First AWS Service Feature to Follow the DuckLabs Deal

Overview

Amazon Aurora PostgreSQL can now query Apache Iceberg and Parquet data in a data lake directly, using an embedded copy of DuckDB, according to the AWS News Blog. InfoQ reported that Aurora PostgreSQL is the first service to receive a user-facing feature powered by DuckDB following Amazon’s acquisition of DuckLabs, the company behind the analytical database. The deal itself was previously reported when Amazon signed the agreement in August.

What the Feature Does

AWS says “DuckDB is now embedded directly within Aurora PostgreSQL, so you can query live operational data (including uncommitted writes) alongside your data lake in a single query,” per the AWS News Blog. The post lists the supported sources as Apache Iceberg tables managed through the AWS Glue Data Catalog, plus Parquet and Iceberg data stored in Amazon S3 and S3 Tables. It also says queries can reach data in catalogs compatible with the Iceberg REST Catalog (IRC) through AWS Glue Data Catalog federation.

The post frames the feature as a replacement for a familiar workaround: “Previously, if your application needed to combine recent transactional data in Aurora with historical records stored in Amazon S3, a common approach was to build reverse ETL pipelines that duplicated data, increased infrastructure costs, and required ongoing engineering effort to keep everything synchronized.” InfoQ adds that query processing stays within Aurora, without additional network hops or ETL pipelines.

How It Is Set Up

According to the AWS News Blog:

• The capability is supported on Aurora PostgreSQL 17, starting with 17.11, and on 18, starting with 18.6.
• Users attach an IAM role with the AuroraAnalytics feature and enable the aurora_analytics extension.
• Users then create foreign tables that point to Iceberg or Parquet data in the data lake. A single IMPORT FOREIGN SCHEMA statement can bulk-create foreign tables for every Iceberg or Parquet table in an AWS Glue Data Catalog database, inferring schemas automatically.
• Aurora applies optimizations such as predicate pushdown and column pruning so that only the relevant data is read, and the aurora_analytics_stat_statements() function reports per-query metrics such as rows scanned, bytes read from Amazon S3, and cache hits.
• For query patterns that need single-digit-millisecond latency, data can be materialized into a native Aurora PostgreSQL table with commands such as CREATE TABLE AS SELECT, INSERT INTO ... SELECT or MERGE INTO. Those commands write data into Aurora, so they run on the writer instance.

The feature is available in all commercial AWS Regions and AWS GovCloud (US) Regions at no additional charge. AWS says customers pay only for the incremental Aurora compute their queries consume and Amazon S3 request costs for reading data lake files.

Context: AWS and DuckLabs

The AWS post states that “DuckLabs, the team that maintains the DuckDB project, recently joined Amazon, and this capability is an example of how the efficiency of DuckDB is being integrated into our services.” In its August announcement on the AWS Big Data Blog, Amazon said it expected the transaction “to close shortly, subject to customary closing conditions,” and that it was “going to look at how we can integrate DuckDB’s performance and simplicity in our other AWS services across data and analytics.”

The same Big Data Blog post said Amazon Quick had processed over 2.5B queries using a custom query engine with DuckDB integrations and optimizations since its October 2025 launch, and that those changes helped reduce average query latency by 30%.

On governance, InfoQ reported that Hannes Mühleisen and Mark Raasveldt, the creators of DuckDB and co-founders of DuckLabs, will continue to lead the team and the open-source project’s technical direction as part of AWS. DuckDB and components of the “Duck Stack” such as DuckLake and Quack will remain free and open source under the MIT licence, with the non-profit DuckDB Foundation continuing its supervision of the projects, InfoQ said. It also noted that the community welcomed the news, though with some scepticism.

What We Don’t Know

• The sources reviewed give no benchmark figures for the Aurora feature, so how it compares with running DuckDB outside the database or with existing federated-query options is not documented there.
• The AWS post does not say which other AWS services will receive DuckDB-powered features, beyond the Big Data Blog’s statement that AWS will look at integrating DuckDB across data and analytics services.
• None of the cited sources gives a date on which the DuckLabs acquisition closed.