false-sharing

Gopium 🌺: Smart Go Structures Optimizer and Manager

traffic-shm (Anna) is a Java based lock free IPC library.

Cache-line false sharing linter for Rust structs

How to solve false sharing problem in C#

A simple high-performance ring buffer handling 10,000,000 messages/sec.

An Evaluation of Different Random Generation Strategies in Java

Proof of Concept demonstrating mechanical sympathy principles from Martin Fowler's article, applied to a low-latency order processing system.

A Simple Demo of How False Sharing Can Affect the Latency & Throughput

Introduction to cache coherence: false sharing, MESI protocol and vectorization

Struct memory layout analyzer for C, C++, Rust, Go and Zig: finds padding waste, false sharing and cache-locality problems, and auto-fixes field order. CLI, CI-ready (SARIF, GitHub Action), LSP and VS Code extension.

to demonstrate the impact of false sharing on performance.

Contador de acessos por URL em logs de CDN, em C com OpenMP. Compara a versão sequencial com estratégias de sincronização paralela (atomic, critical, locks por bucket e atomic com padding contra false sharing) sobre uma hash table própria, com benchmark de escalabilidade por número de threads.

Cacheline false sharing benchmarks in C++, Go, and Linux perf-c2c profiling scripts.

Custom multithreaded C++17 memory allocator: thread-local lock-free pools over a locked central heap, global new/delete override, constexpr compile-time fragmentation detection, and a real-time ImGui visualizer. ~3-4x faster than malloc.

Twelve small benchmarks behind the video "Every Reason Your Code Is Slow, Explained in 14 Minutes": cache order, pointer chasing, false sharing, branches, SIMD, allocation, locks, syscalls.

Parallel Programming HW1 (UoA DI): four pthread exercises — Monte Carlo π, mutex vs C11 atomics, false-sharing demo with cache-line padding, sorted linked list with custom read-write locks. Each ships data sweeps + plots. Solo project.

Lock-free SPSC queue in modern C++ with a benchmarking suite comparing lock-free and mutex queues across workloads.

Interactive C++ simulator for cache-line false sharing, coherence traffic, stalled CPU time, and counter-layout tradeoffs.

C benchmark demonstrating CPU cache false sharing via padded vs unpadded concurrent counters, driven by a Python LangChain agent that sweeps thread counts and writes an automated analysis

Some experiments, tests, benchmarks, etc.