52% More Throughput. 1.5x Better Value. The Azure SQL DB Hyperscale Advantage | Data Exposed
Ravinder Singh and Anna Hoffman examine the benchmark’s throughput and cost-efficiency results in this Data Exposed episode, focusing on the comparison as an input to cloud database selection.
Benchmark scope and results
Principled Technologies tested Azure SQL Database Hyperscale against Amazon Aurora PostgreSQL I/O-Optimized using an I/O-intensive HammerDB workload. The study covered two vCore counts and multiple concurrency loads.
Across those configurations, Azure SQL Database Hyperscale delivered similar or stronger database performance than Amazon Aurora PostgreSQL. The most notable results were at 16 vCores, where Hyperscale delivered:
- Up to 52% higher performance
- Up to 51% better performance per dollar, approximately 1.5 times the value
These figures describe results from the tested workload and configurations, rather than a universal performance advantage for every application. For cloud selection, the relevant comparison is both workload performance and the cost of delivering that performance.
The Principled Technologies benchmark study provides the source for the comparison and reported results.