Amazon QuickSight Adds S3 Tables for AI-Ready Analytics

Amazon QuickSight has introduced Amazon S3 Tables — built on Apache Iceberg — as a new native data source, allowing customers to directly query and visualize data stored in S3 table buckets without any intermediate data layers. The update represents a significant step in AWS's broader strategy to simplify modern data architectures and bring near real-time analytics closer to where data already lives.

This integration eliminates a long-standing pain point for data engineers and analysts who previously had to move, transform, or duplicate data across multiple services before it could be visualized. Now, with a direct connection from QuickSight to S3 Tables, organizations can dramatically reduce pipeline complexity while accelerating time-to-insight.

Key Facts at a Glance

• New data source: Amazon QuickSight now natively connects to Amazon S3 Tables stored in S3 table buckets
• Format: S3 Tables are built on the open-source Apache Iceberg table format
• No middleware required: Queries run directly against Iceberg tables — no ETL pipelines, no intermediate data layers
• Near real-time analytics: The architecture supports low-latency querying for dashboards and visualizations
• Architecture simplification: Reduces the number of services in a typical analytics stack by removing redundant transformation steps
• Target users: Data engineers, analytics teams, and business intelligence professionals on AWS

What Are Amazon S3 Tables and Why Do They Matter?

Amazon S3 Tables represent a managed implementation of Apache Iceberg tables within Amazon S3. Unlike traditional S3 objects, which store flat files requiring external catalog and query engines to interpret, S3 Tables bring structured, queryable table semantics directly into the storage layer.

Apache Iceberg itself has emerged as one of the most popular open table formats in the data lakehouse ecosystem, competing with Delta Lake (backed by Databricks) and Apache Hudi. Iceberg's appeal lies in its support for ACID transactions, schema evolution, time travel, and partition evolution — features that make large-scale analytical datasets far more manageable.

By embedding Iceberg tables natively within S3, AWS effectively collapses what used to require separate catalog services (like AWS Glue Data Catalog), query engines (like Amazon Athena), and transformation layers into a more streamlined architecture. The addition of QuickSight as a direct consumer of this data closes the loop from storage to visualization.

How the Integration Eliminates Pipeline Complexity

Traditional analytics workflows on AWS typically follow a multi-hop architecture. Raw data lands in S3, gets cataloged by Glue, transformed by Glue ETL jobs or AWS Lambda functions, queried through Athena or Amazon Redshift Spectrum, and finally visualized in QuickSight. Each hop introduces latency, cost, and maintenance overhead.

The new S3 Tables integration in QuickSight removes several of these intermediate steps. Analysts can now point QuickSight directly at an S3 table bucket, select the relevant Iceberg tables, and begin building dashboards immediately. This is particularly impactful for organizations managing petabyte-scale data lakes where every additional data movement operation carries significant cost implications.

Compared to the previous workflow — which could require 4 to 6 distinct AWS services working in concert — the new approach can reduce the analytics pipeline to as few as 2 components: S3 Tables for storage and QuickSight for visualization.

Near Real-Time Analytics Become More Accessible

One of the most compelling aspects of this update is the near real-time analytics capability it unlocks. Because Apache Iceberg supports incremental data ingestion and snapshot-based reads, QuickSight can now surface fresh data without waiting for batch ETL jobs to complete.

This matters enormously for use cases like:

• Operational dashboards tracking live KPIs across e-commerce, logistics, or SaaS platforms
• Financial monitoring where delays of even minutes can impact decision-making
• IoT analytics processing continuous streams of sensor data stored in Iceberg format
• Customer behavior analysis requiring up-to-the-hour session and engagement metrics
• Supply chain visibility where inventory and shipment data changes rapidly

Previously, achieving near real-time visualization on AWS often required Amazon Redshift or Amazon OpenSearch Service as an intermediary, adding both cost and architectural complexity. The S3 Tables integration offers a lighter-weight alternative for many common analytics scenarios.

Industry Context: The Data Lakehouse Convergence Continues

This announcement fits squarely within the broader data lakehouse trend that has been reshaping the analytics industry since Databricks popularized the concept around 2020. The core idea — combining the flexibility and cost-efficiency of data lakes with the performance and governance of data warehouses — has become the dominant architectural paradigm.

AWS has been steadily building out its lakehouse capabilities over the past 3 years. The launch of S3 Tables in late 2024 was a pivotal moment, signaling that AWS was ready to compete more directly with Databricks' Unity Catalog and Delta Lake ecosystem. Adding QuickSight as a first-class consumer of S3 Tables is the logical next step.

Meanwhile, the competitive landscape remains intense. Databricks recently enhanced its own BI capabilities with the acquisition-powered Databricks SQL and AI/BI Dashboards. Snowflake continues to push its Iceberg Tables support as a key differentiator. Google BigQuery has similarly added Iceberg integration through BigLake. AWS's move ensures QuickSight remains competitive in this rapidly evolving space.

The broader trend is clear: every major cloud provider is racing to eliminate friction between data storage and data visualization, and open table formats like Iceberg are becoming the common language of this convergence.

What This Means for Developers and Data Teams

For practitioners, this update has several immediate practical implications:

• Reduced infrastructure costs: Fewer services in the pipeline means fewer resources to provision, monitor, and pay for
• Faster prototyping: Data teams can build proof-of-concept dashboards directly on raw Iceberg tables without setting up transformation pipelines first
• Simplified governance: With fewer data copies and movement steps, data lineage becomes easier to track and compliance requirements easier to meet
• Lower skill barriers: Business analysts can access Iceberg data through QuickSight's visual interface without needing SQL expertise in Athena or Redshift

Data engineers should note, however, that this integration does not replace all use cases for a full-featured query engine. Complex analytical workloads involving heavy joins, window functions, or machine learning feature engineering will still benefit from dedicated engines like Athena, Redshift, or Amazon EMR. The S3 Tables data source in QuickSight is best suited for direct visualization and lightweight aggregation scenarios.

Organizations already invested in Apache Iceberg as their table format standard will find the transition particularly smooth. Those currently using Delta Lake or Hudi may need to evaluate format migration strategies if they want to take advantage of this specific integration.

Looking Ahead: AI-Ready Analytics on AWS

The phrase 'AI-ready analytics' in AWS's messaging around this feature is deliberate and forward-looking. As organizations increasingly look to feed analytics data into machine learning models and generative AI applications, the ability to maintain clean, queryable, well-governed datasets becomes critical.

S3 Tables' Iceberg foundation provides built-in support for schema evolution and time travel — features that are essential for ML reproducibility and data versioning. By making this data directly accessible in QuickSight, AWS is positioning the service not just as a BI tool, but as part of a broader AI-ready data platform.

Looking further ahead, we can expect AWS to deepen this integration in several ways. Tighter coupling with Amazon SageMaker for direct model training on Iceberg data is a likely next step. Integration with Amazon Q, AWS's generative AI assistant, could enable natural language querying of S3 Tables directly within QuickSight dashboards.

The trajectory is clear: AWS wants to make the journey from raw data lake to AI-powered insight as frictionless as possible. The S3 Tables integration in QuickSight is an important milestone on that path — one that brings real, measurable benefits to data teams today while laying the groundwork for more intelligent analytics tomorrow.

For organizations evaluating their 2025 analytics architecture, this update makes a compelling case for standardizing on Apache Iceberg and leaning into the native AWS toolchain. The days of building sprawling, multi-service analytics pipelines may finally be numbered.