Lock-free SPSC Ring Buffer

This project is designed for high-throughput, low-latency real-time computing scenarios—such as audio signal processing and high-performance network packet forwarding—featuring zero memory allocation, cache-line alignment, and extreme copy optimizations.

Furthermore, adhering to modern C++ engineering practices, the project fully integrates Google Test for unit testing, Google Benchmark for micro-benchmarking, and a GitHub Actions CI pipeline supporting multiple platforms and compilers.

---

Lock-free SPSC Ring Buffer Models

SPSC Ring Buffer Schematic Diagram:

Data Race wihout Happens-Before:

pictures from CppCon2023: Single Producer Single Consumer Lock-free FIFO From the Ground Up by Charles Frasch

---

Documents

To understand the lock-free mechanics and low-level optimizations implemented in this project, check out the engineering docs:

• Concurrency & Atomic - analyse the danger of data race and discuss how to establish lock-free safey boundaries by using happens-before
• Memory Order - discuss how to move away from costly seq_cst and leverage acquire, release semantics to squeeze out maximum hardware performance
• Hardware & Cache - explain the chache line mechanism and how to make optimizations by using alignas to mitigate false sharing
• Data Copy & Safety - compile-time assertions(static_assert) and how to maximize contiguous memory throughput by using memcpy