C++ Atomic Operations: std::atomic & High-Performance Lock-Free Architecture

In ultra-high-performance systems (high-frequency trading, real-time audio, game rendering engines), the latency of acquiring a mutex lock is unacceptable. C++ provides Atomic Operations via <atomic> that execute indivisibly at the CPU hardware level.

Lock-Free Counter with std::atomic<int>

#include <iostream>
#include <atomic>
#include <thread>
#include <vector>

// Lock-free atomic integer! Zero mutexes needed!
std::atomic<int> atomicCounter{0};

void worker() {
    for (int i = 0; i < 50000; ++i) {
        atomicCounter.fetch_add(1, std::memory_order_relaxed); // Hardware atomic increment
    }
}

int main() {
    std::vector<std::jthread> threads;
    for (int i = 0; i < 4; ++i) {
        threads.emplace_back(worker);
    }
    threads.clear(); // Waits for completion

    std::cout << "Atomic Counter Result: " << atomicCounter.load() << "\n"; // 200,000!
    std::cout << "Is hardware lock-free: " << std::boolalpha << atomicCounter.is_lock_free() << "\n";

    return 0;
}