THE PERFORMANCE COST OF RWLOCK IN A READ-HEAVY WORKLOAD

Summary from the TLDR Dev newsletter

A Rust benchmark found that replacing 16,384 per-item RwLock acquisitions with one Crossbeam epoch guard and atomic pointer loads raised read throughput from 15.93k to 229.07k operations per second, while a simple lock still worked well around an infrequently changed collection.