Lobsters is a community-driven platform for sharing and discussing links to articles, tutorials, and projects related to technology and programming. Readers can learn about a wide range of topics, from software development and system administration to cybersecurity and artificial intelligence. With user submissions, comments, and voting, Lobsters provides a platform for collaborative learning and knowledge sharing among technology enthusiasts.

Lobsters

A hands-on investigation into why RwLock performed ~5× worse than Mutex for a read-heavy LRU tensor cache in Rust. The root cause is cache-line ping-pong: even a read lock requires an atomic increment of an internal reader counter, causing every core to fight for exclusive ownership of that cache line. For very short critical sections (nanosecond-scale hashmap lookups), this contention overhead dominates actual work time. Key takeaways: prefer Mutex for tiny critical sections, use sharding to reduce contention, and always profile with tools like perf or flamegraph before assuming RwLock is faster for read-heavy workloads.

Read Locks Are Not Your Friends