Servers ability to serve concurrent requests at higher throughput can reduce the resources footprint of a service.

This talk reviews approaches to increase Apache Sling throughput and presents practical experiments with a multi-threaded server model backed by OpenJDK project Loom's virtual threads.