Key Features
- Pool allocator for orders: fixed-size blocks with the free list stored inside the free memory, so allocating and freeing an order is O(1) with no heap fragmentation.
- Linear allocator as scratch space for each incoming order message, reset in O(1) once the message is processed.
- Price-time-priority matching: buy orders kept highest-price-first and sell orders lowest-price-first; an incoming order fills against the best opposite price while prices cross.
- Benchmarked 1M allocate/free operations (median of 7 runs, g++ -O2 on Linux): the pool allocator is ~2.5x and the linear allocator ~15x faster than glibc new/delete.
- Stack (LIFO) and general-purpose Free-List (first-fit with coalescing) allocators share the same interface for side-by-side comparison.