Posted On: Mar 26, 2024
AWS announces the availability of Reserved Instances for Amazon Aurora Optimized Reads for Aurora PostgreSQL. Optimized Reads is a price-performance capability available on r6gd and r6id instances that delivers up to 8x improved query latency and up to 30% cost savings compared to instances without it, for applications with large datasets that exceed the memory capacity of a database instance. Reserved Instances are appropriate for steady-state database workloads and offer significant savings over On-Demand Instances improving these savings. 1-year Reserved Instances provide discounts up to 27% and 3-year Reserved Instances provide discounts of up to 47% over On-Demand Instances. Reserved Instances provide flexibility across instance sizes and cluster configuration options.
Optimized Reads use local NVMe-based SSD block-level storage available in r6gd and r6id instances to improve query latency of applications with data sets exceeding the memory capacity. It includes two features: tiered caching and temporary objects. Temporary objects achieve faster query processing by placing temporary tables on local storage, improving the performance of queries involving sorts, hash aggregations, and other data-intensive operations. Tiered caching—available only in I/O-Optimized—automatically stores data evicted from the in-memory database buffer cache onto local storage to speed up subsequent data accesses. Tiered caching accelerates the query throughput of pgvector-based similarity searches by up to 9x, improving the performance of Generative AI applications. To learn more about Optimized Reads Reserved Instances visit our pricing page, and documentation.