Scalable and Low Latency Lock-free Data Structures in C++

Explore scalable and low latency lock-free data structures in C++ through this comprehensive conference talk from CppCon 2022. Delve into the challenges of developing high-performance data structures for multi-threaded programs handling millions of operations per second. Learn about lock-free approaches, cache-conscious data structures, and their impact on modern hardware performance. Examine various lock-free hash table variations and cache-conscious trees, discussing their optimal use cases and limitations. Focus on addressing worst-case scenarios and tail latency issues in large, busy systems. Follow a step-by-step C++ implementation of Hash Trie, a hybrid lock-free cache-conscious data structure, and compare its performance through microbenchmarks. Gain insights into x86-64 memory ordering, cache hierarchy, operating system preemption, and their application in implementing fast data structures. Discover the intricacies of data structure benchmarking, including key distribution, latency vs. throughput, and worst-case analysis. Access an open-source lock-free cache-conscious Hash Trie implementation and expand your knowledge of advanced container solutions beyond standard containers and locking mechanisms.