yinhe

A high-performance order-matching engine written in C++17.

• Orderbook — Price-time priority matching engine using std::map (sorted bid/ask levels) and std::list (FIFO per level). Supports Good-To-Cancel (GTC), Fill-Or-Kill (FOK), and Fill-And-Kill order types.
• Lock-free SPSC logger — Trades are logged asynchronously via a single-producer/single-consumer ring buffer. The consumer thread spin-polls the queue, avoiding condition_variable syscall overhead on the hot path.
• O(1) cancel — Orders are indexed by ID in an unordered_map pointing directly into the level's linked list, so cancellation is a constant-time erase.

Monte Carlo benchmark: 10 runs per scenario, compiled with -O3 -march=native. Each run uses a fresh orderbook.

Measured on Apple M-series. Logger enabled (realistic path including SPSC push).

# Debug
cmake -B build && cmake --build build

# Release (benchmarks)
cmake -B build-release -DCMAKE_BUILD_TYPE=Release \
  -DCMAKE_CXX_FLAGS="-O3 -std=c++17 -DNDEBUG -march=native"
cmake --build build-release

# Tests
./build/bin/test_orderbook
./build/bin/test_orderbook_stress

# Benchmark
./build-release/bin/bench_orderbook

src/
  engine/
    orderbook.{hpp,cpp}   — core matching engine
    order.{hpp,cpp}        — order value type
    levelInfo.hpp          — price level aggregation
    tradeUtils/trade.hpp   — trade result type
  common/
    orderLog.hpp           — async SPSC logger
    SPSCQueue.hpp          — lock-free ring buffer
    types.hpp, enums.hpp   — shared type aliases and enums
  tests/
    test_orderbook.cpp     — unit tests
    test_orderbook_stress.cpp — stress / edge-case tests
    bench_orderbook.cpp    — Monte Carlo performance benchmark
  main.cpp                 — demo entry point