Next-Generation Performance with AWS Graviton
Since its debut in 2013, Amazon Redshift has continually pushed the boundaries of cloud data warehousing, delivering enterprise-grade analytics at a fraction of on-premises costs. Each architectural leap—from dense compute nodes to RA3 instances, and from provisioned clusters to serverless options—has steadily driven down query costs and accelerated performance. Now, with the introduction of Redshift RG instances, AWS introduces a new instance family powered by AWS Graviton processors. These instances deliver up to 2.2 times faster data warehouse performance compared to RA3 instances, while slashing the price per vCPU by 30%. This combination of speed and cost efficiency makes RG instances an ideal foundation for modern analytics and AI-driven workloads.
Key Performance Gains
RG instances are optimized for both traditional SQL queries and emerging AI agent scenarios. For example, low‑latency business intelligence (BI) dashboards, ETL pipelines, and near‑real‑time analytics see dramatic responsiveness improvements. Additionally, autonomous AI agents—which query data warehouses at volumes far exceeding human usage—benefit from the higher throughput and lower latency. In March 2026, Redshift had already boosted new query performance by up to 7 times for BI and ETL tasks; RG instances further amplify these gains.
Unified Analytics Across Data Warehouse and Data Lake
Organizations increasingly rely on both structured data warehouse tables and cost‑effective data lakes built on Amazon S3. The new RG instances feature an integrated data lake query engine that lets you run SQL analytics across both environments from a single engine. This integration delivers up to 2.4 times faster performance on Apache Iceberg tables and up to 1.5 times faster on Apache Parquet files compared to RA3 instances. By eliminating the need to shuttle data between separate systems, you simplify operations, reduce total analytical costs, and speed time‑to‑insight.
How It Helps
- Single engine for querying warehouse tables and Amazon S3 data lakes.
- Reduced complexity – no need to maintain separate query engines.
- Lower costs – combined data warehouse + data lake workloads become more economical.
To estimate savings for your specific workloads, use the AWS Pricing Calculator.
RG Instances vs RA3 Instances: A Quick Comparison
When migrating from current RA3 instances, you can expect both a performance boost and a better price‑to‑performance ratio. Here’s how the two instance families compare for common production workloads:
- ra3.xlplus → rg.xlarge (4 vCPUs, 32 GB RAM) – ideal for small clusters used in departmental analytics.
- ra3.4xlarge → rg.4xlarge (16 vCPUs, 128 GB RAM) – a 1.33:1 improvement in both vCPUs and memory, designed for standard production workloads with medium data volumes.
This straightforward migration path helps you start reaping the benefits of Graviton‑powered performance with minimal disruption.
Designed for Modern Analytics and AI Agent Workloads
As data volumes grow and analytics requirements evolve, two key trends emerge: the blending of warehouse and data lake storage, and the rise of AI agents that query data at machine scale. RG instances are purpose‑built for these new realities. They handle the high concurrency and low latency demanded by:
- Business intelligence dashboards with sub‑second response times.
- ETL pipelines that need to process large volumes of data quickly.
- Near‑real‑time analytics applications.
- Autonomous, goal‑seeking AI agents that execute thousands of queries per second.
By combining faster compute with an integrated data lake query engine, Redshift RG instances ensure that your analytics infrastructure scales alongside your business—without driving up costs.
How to Get Started with RG Instances
You can launch new Redshift clusters using RG instances or migrate your existing RA3‑based clusters directly from the AWS Management Console, AWS CLI, or AWS API. The integrated data lake query engine is enabled by default, so you start benefiting immediately. There’s no additional configuration required for querying data in Amazon S3 alongside your warehouse tables. For detailed steps, refer to the Amazon Redshift Management Guide.
The Future of Cloud Data Warehousing
With RG instances, Amazon Redshift takes another significant step toward making data analytics faster, more affordable, and simpler to manage. By leveraging AWS Graviton processors and embedding a query engine capable of spanning warehouse and lake, AWS addresses two critical pain points for modern organizations: the need for better price‑performance and the desire to unify previously siloed data stores. Whether you’re running conventional BI workloads or pioneering AI‑driven automation, RG instances provide the foundation to do more with your data—at a lower cost.