## My Conclusion: Summation of a Vector in three Variants

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

Views: 31748

## Multithreaded: Summation with Minimal Synchronization

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 from each other as possible and therefore reduce the synchronization overhead.

Views: 17168

## Multithreaded: Summation of a Vector

My goal is, 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.

Views: 38692

## Single Threaded: Summation of a Vector

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 the reference number. In further posts, I discuss atomics, locks, and thread-local data.

Tags: atomics, lock
Views: 21368

## Thread-Safe Initialization of a Singleton

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.

Views: 276740

## Ongoing Optimization: Relaxed Semantic with CppMem

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

Views: 8769

## Ongoing Optimization: A Data Race with CppMem

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?

Views: 8854

## Ongoing Optimization: Acquire-Release Semantic with CppMem

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.

Views: 10053

## Ongoing Optimization: Sequential Consistency with CppMem

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.

Views: 10794

## Ongoing Optimization: Locks and Volatile with CppMem

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

Views: 11021

### Subscribe to the newsletter (+ pdf bundle)

 Name Email Please enable the javascript to submit this form

### Visitors

Today 7315

Yesterday 7224

Week 29314

Month 34579

All 6892788

Currently are 255 guests and no members online

• #### Template Instantiation

if you're testin and have a single file, you can just \$ make mainMyClass

• #### Performance of the Parallel STL Algorithms

the global variable size is conflicts with std::size introduced from c++17.

• #### C++ Core Guidelines: Rules for Enumerations

How about a recommendation on the use of boolean typed enums instead of boolean values as parameters ...

• #### Performance of the Parallel STL Algorithms

Dude the name is GNU/Linux not Linux. You could have done both GCC and VC on Windows for more fair ...

• #### Atomic Smart Pointers

Maybe I'm missing something. It seem to me that a better way to describe the problem is that shared_ptr's ...