Understanding the Performance Implications of Storage-Disaggregated Databases

Storage-compute disaggregation in databases decouples compute from storage, allowing independent scaling and shared storage use. This architecture is gaining traction in cloud environments like Amazon Aurora and Google AlloyDB. However, it introduces performance challenges such as cache misses and replication lags. Optimizations like buffering and multi-versioning can mitigate these issues. Despite some unresolved performance questions, disaggregated databases offer significant benefits, including fault tolerance and easier sharding. The study provides open-source implementation and insights into performance trade-offs.