What is AWS Aurora?

• Aurora is part of Amazon RDS.
• It supports MySQL and PostgreSQL relational databases.
• It offers cost-effectiveness with auto-scaling capabilities.
• It provides simplicity in setup and operation.
• Aurora supports clustering replication.

What is Aurora Limitless?

• Aurora Limitless supports PostgreSQL-compatible Aurora databases.
• It allows scaling beyond the limits of an Aurora database instance.
• It enables horizontal scaling of write throughput for increased storage capacity.
• It ensures transactional consistency by distributing data and query load across multiple Serverless instances.
• Aurora Limitless features auto-scaling, eliminating the need for manual provisioning.
• Additional capacity for writers and readers is provided through the Aurora cluster.
• Distributed query planning and transaction management facilitate transparent scaling complexity.

What is distributed query planning on scaling complexity transparently?

• Distributed query planning optimizes queries.
• It allows retrieval of data stored across multiple databases.
• It enables optimal parallelization of query execution while maintaining transactional consistency.
• It behaves as a single logical instance, optimizing performance and resource utilization.

Benefits:

• Supports high write throughput and storage beyond a single instance's capabilities.
• Enables horizontal scaling for unpredictable and rapidly growing workloads.
• Automates sharding and replication processes.

What is transaction management on scaling complexity transparently?

• involves coordinating operations across distributed databases.
• ensures data integrity by addressing issues related to parallelized operations, concurrency, replication, and distributed queries.
• coordinates transactions across scaled-out databases, allowing seamless horizontal scaling.

Sharded tables & Reference tables

AWS Aurora Limitless Database has two types of tables: Sharded tables & Reference tables.

What are Sharded tables?

• Sharded tables are distributed across multiple database shards.
• Data is split based on designated columns (shard keys) and stored in sharded tables.
• Sharded tables are used for extremely large tables that need to be horizontally partitioned across instances.
• The main benefit of sharded tables is scaling and achieving high throughput.

What are Reference tables?

• Reference tables contain data that is present on every shard.
• They allow join queries to work faster by reducing unnecessary data movement across shards.
• Reference tables are typically used for infrequently modified reference data, such as zip codes.

Use cases of using Aurora:

• Real-time analytics on large datasets.
• High-volume transaction processing.
• Ideal for managing massive datasets and heavy database usage.

Example 1: GSR

• GSR, one of the largest Crypto Market Makers.
• Handles more than 1.1 million trades per day.
• Utilizes Aurora for high performance and availability at a global scale.

Example 2: Magnetic Asia

• Magnetic Asia operates a live streaming events platform called Total Streaming, with up to 50,000 attendees.
• Their virtual ticketing solution, Total Ticketing, relies on Aurora.
• Aurora intelligently scales the database based on workload requirements.

Reference

Announcing Amazon Aurora Limitless Database

Join the preview of Amazon Aurora Limitless Database

Embracing the Future of Database Management: Amazon Aurora Limitless Database

From 0 to Limitless, the 15-year odyssey of Amazon relational database product line

Nuts and Bolts of Transaction Processing

GSR Scales Fast on AWS to Become One of the Largest Crypto Market Makers

Case Studies for Financial Services

Case Study: Magnetic Asia keeps entertainment alive with live streaming services on AWS