Amazon RDS

Improved vector similarity search for generative AI applications with 20x improved queries per second vs pgvector_IVFFLAT

Automatically back up database and transaction logs with retention up to 35 days, allowing restoration to any second during retention period

User-initiated backups stored in Amazon S3, kept until explicitly deleted, with ability to create new instances from snapshots

Record and audit changes to DB instance configuration including parameter groups, subnet groups, snapshots, and security groups

Support for eight relational database engines including PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, Db2, and Aurora

Safer, simpler database updates with zero data loss by creating staging environment that mirrors production with logical replication

Managed experience for applications requiring customization of underlying operating system and database environment

Deploy fully managed database instances in on-premises environments for hybrid deployments

Quick, efficient cloning of entire multiterabyte database clusters in minutes without storage charges for initial clones

Automatically replicate database snapshots and transaction logs across regions for disaster recovery

Replicate data across regions with free data transfer for replication in Multi-AZ deployments

Achieve high availability with SLA up to 99.95% through synchronous replication to standby instances in different Availability Zones

Deploy one primary and two readable standby instances across three AZs with failovers in under 35 seconds and 2x faster transaction commit latency

Leverage Graviton3-based instances for optimized performance and cost efficiency

Model, provision, and manage RDS resources across AWS accounts and Regions using CloudFormation templates

Automatically replicate transactional data to Amazon Redshift or Amazon SageMaker for analytics and ML without building data pipelines

Automates undifferentiated tasks including software patching, backing up, provisioning, and scheduling maintenance jobs

Centrally manage and stagger automatic minor version upgrades across multiple AWS resources and accounts

Granular control and fine-tuning of database configurations

Create production-ready relational databases in minutes using AWS Management Console, CLI, SDKs, or API calls

Optional control over when and if database instances are patched

Easy-to-use tool that helps quickly detect and diagnose database performance problems

Consolidates logs and metrics from applications, databases, and operating systems with pre-built dashboards and automated telemetry collection

Support for Amazon CloudWatch metrics for database instances at no additional charge with unified console view

Access to over 50 CPU, memory, file system, and disk I/O metrics for database instances

ML-powered cloud operations service that automatically identifies root causes of performance issues and provides intelligent recommendations

Receive email or SMS notifications for over 40 different database events through Amazon SNS

Provide up to 2x faster query processing for complex queries using temporary tables, sorts, hash aggregations, and joins

Improve write transaction throughput by up to 2x using AWS Nitro System Torn Write Prevention, especially useful for write-intensive workloads

Add managed in-memory caching service directly from RDS console to accelerate database and application performance

Provides price predictability for database workloads

Make applications more scalable, resilient, and secure with connection pooling

Flexible pricing model offering savings in exchange for 1-year commitment to specific usage (measured in $/hour)

AI-powered analysis of configuration and usage metrics providing recommendations for database engine versions, storage, instance types, and networking

Automatically scale up to hundreds of thousands of transactions in a fraction of a second for unpredictable workloads

Support for high-performance EBS io2 Block Express storage for demanding workloads

Cost-effective SSD storage option with 3 IOPS per provisioned GB baseline and burst up to 3,000 IOPS, ranging from 20 GiB to 64 TiB

High-performance SSD storage for I/O-intensive workloads, scaling from 1,000 to 256,000 IOPS with 100 GiB to 64 TiB capacity

Continue using MySQL and PostgreSQL major versions after community end-of-life with critical security fixes and bug patches for up to 3 years