Aurora Limitless – Distributed PostgreSQL at Scale

Amazon Aurora Limitless Database: Scaling to Millions of Transactions Per Second

Aurora Limitless Database, launched in 2024, is now production-ready for the most demanding workloads. It automatically shards your data across multiple Aurora instances while maintaining a single-database experience. Here’s what you need to know.

Aurora Limitless vs Traditional Aurora

How It Works

Aurora Limitless uses distributed routers and shards:

-- Create a sharded table
CREATE TABLE orders (
    order_id BIGINT PRIMARY KEY,
    customer_id BIGINT,
    order_date TIMESTAMP,
    total DECIMAL(10,2)
) WITH (
    limitless = true,
    shard_key = customer_id  -- Distribution key
);

-- Reference tables replicated to all shards
CREATE TABLE products (
    product_id INT PRIMARY KEY,
    name VARCHAR(255),
    price DECIMAL(10,2)
) WITH (
    limitless = true,
    collocated = true  -- Full copy on each shard
);

-- Queries work normally
SELECT * FROM orders
WHERE customer_id = 12345;  -- Routes to single shard

SELECT customer_id, SUM(total)
FROM orders
GROUP BY customer_id;  -- Distributed query

When to Use Aurora Limitless

Migration Path

# 1. Create Limitless cluster
aws rds create-db-cluster \
    --db-cluster-identifier my-limitless \
    --engine aurora-postgresql \
    --engine-mode limitless \
    --master-username admin \
    --master-user-password secret

# 2. Use DMS to migrate with minimal downtime
# 3. Update application connection string
# 4. Add shard keys to tables

Aurora Limitless removes the biggest scaling bottleneck in relational databases. If you’ve been considering NoSQL for scale, take another look at Aurora.