After I've calculated in three different ways the sum of a std::vector I want to draw my conclusions.

Until now I've used two strategies for the summation of a std::vector. First I did the whole math in one thread (Single Threaded: Summation of a vector); second multiple threads shared the same variable for the result (Multithreaded: Summation of a vector). In particular the second strategy was extremely naive. In this post I will apply my knowledge of both posts. My goal is it that the thread will perform their summation as independent form each other as possible and therefore reduce the synchronization overhead. 

My goal is it to sum up all elements of a vector. I used in the last post a single thread. In this post I use multiple threads and therefore the full power of my PC. The addition will be done on a shared variable. What at first glance seems like a good idea is a very naive strategy. The synchronization overhead of the summation variable is higher than the performance benefit of my four or two cores.

What is the fastest way to add the elements of a std::vector?. A question which I will pursue in the next posts. I use the single threaded addition as reference number. In further posts I discuss atomics, locks, and thread local data.

There are a lot of issues with the singleton pattern. I'm totally aware of that. But the singleton pattern is an ideal use case for a variable, which has only to be initialized in a thread safe way. From that point on you can use it without synchronization. So in this post I discuss different ways to initialize a singleton in a multithreading environment. You get the performance numbers and can reason about your uses cases for the thread safe initialization of a variable.

With the relaxed semantic, we have no synchronisations and ordering constraints on atomic operations.

But we can do better and further improve the acquire-release semantic of the last post. Why should x be an atomic? There is no reason. That was my first, but incorrect assumption. See why?

With the acquire-releae semantic, we break the sequential consistency. In the acquire-release semantic the synchronization takes place between atomic operations on the same atomic and not between threads.

With atomic data types you can tailor your program to your needs and therefore optimize it. But now we are in the domain of the multithreading experts.

The easiest way to solve the undefined behaviour in the post Ongoing Optimization: Unsynchronized access is, to use a lock.