AWS recently updated Amazon Aurora Serverless with a new platform version that improves scaling behavior and runtime efficiency. According to the cloud provider, capacity can now scale up about 45% faster during demand spikes, and database performance improves by up to 30% through better resource scheduling and workload-aware scaling decisions.
Platform version 4 introduces improved runtime efficiency and a smarter scaling algorithm, enabling faster capacity scaling and higher database performance for Aurora Serverless clusters.
Generally available for four years, Amazon Aurora Serverless is an on-demand, automatically scaling configuration for Aurora that adjusts database capacity based on application demand. Supporting MySQL and PostgreSQL, Aurora Serverless scales down to zero capacity when not in use, adjusting capacity in small increments of 0.5 ACUs to match workload changes.
A HammerDB TPROC-C benchmark was used to measure performance across three Aurora Serverless platform versions with 1,024 virtual users. The results compare New Orders per Minute (NOPM) for each version and report the percentage change relative to the preceding platform version. Jiaming Yan, senior software development engineer at AWS, Ashok Kurakula, software development engineer III at AWS, and Nashad Safa, senior software engineer at AWS, write:
For both database engines supported on serverless, Aurora MySQL and Aurora PostgreSQL, platform version 4 delivers 27-34% higher NOPM compared to platform version 3.
Source: AWS blog
A Sysbench benchmark compared three Aurora Serverless clusters running platform versions 2, 3, and 4, with identical capacity settings ranging from 0.5 to 256 ACUs and faster scaling enabled. After loading 250 tables totaling 16 GB, a read-heavy workload executed 50 million queries with 512 threads. Evaluating the CloudWatch ServerlessDatabaseCapacity metric to compare capacity across versions, Yan, Kurakula, and Safa conclude:
Platform version 4 delivers 27% faster completion with 28% lower cost than platform version 3, and 41% faster completion with 42% lower cost than platform version 2.
Corey Quinn, chief cloud economist at The Duckbill Group, writes in his newsletter:
Aurora Serverless scaling up 45% faster and down to zero, which is coincidentally where my enthusiasm for "serverless" databases that still bill in ACU fractions tends to land. Genuine improvements at no extra charge, though, which either means competitive pressure is working or someone in Seattle accidentally approved the wrong PRFAQ.
According to the announcement, the enhanced scaling algorithm benefits workloads where multiple tasks compete for resources, such as busy web applications and API services. Pini Dibask, principal database solutions architect at AWS, comments on LinkedIn:
What makes this especially interesting (...) is that Database Savings Plans which was announced at re:Invent 2025 offers Aurora Serverless the highest discount of any AWS database service (35% discount), combined with the performance gains of Platform Version 4 (which should translate into lower ACU consumption), the cost equation for Aurora Serverless has fundamentally shifted.
The update to the latest platform version applies automatically to new clusters and is available to existing clusters via the ServerlessV2PlatformVersion parameter.