
Presto is a popular analytics query engine that provides interactive analytics on configurable data sources. While this architecture provides flexibility, query performance can be dramatically affected by remote storage latency variance. Alluxio is a virtual file system that can provide transparent caching for remote storage systems.
When the two projects joined together, we had an analytical query engine that can query data from anywhere, while caching the hot dataset locally. This architecture is used in many large companies like Meta, Uber, ByteDance, etc, providing flexibility and speed at the same time.