openmp

1. OpenMP线程数设置 (1) 查看核心数： (2) OpenMP获取CPU核心数： omp_get_num_procs() 函数会返回机器的核心数 (3)OpenMP设置线程数： #pragma omp parallel for num_threads(2*numProcs-1) 2. VS开启OpenMP支持 文章标 转载请标明出处: OpenMP设置线程数及开启方法 文章来源: OpenMP设置线程数及开启方法

说明：这部分内容比较基础，主要是分析几个容易混淆的OpenMP函数，加以理解。 （1）并行区域数量的确定： 在这里，先回顾一下OpenMP的parallel并行区域线程数量的确定，对于一个并行区域，有一个team的线程去执行，那么该分配多少个线程去执行呢？ OpenMP的遇到parallel指令后创建的线程team的数量由如下过程决定： 1. if子句的结果 5. 编译器默认实现（一般而言，默认实现的是总线程数等于处理器的核心数） （ http://blog.csdn.net/gengshenghong/article/details/6956878 查看更多信息） 2、3、4优先级依次降低的，也就是前面的设置可以覆盖后面的设置，当然也是相对而言，num_threads子句只会影响当前的并行区域，而omp_set_num_threads对OMP_NUM_THREADS环境变量的覆盖是在整个程序运行期间全局的。 （2）几个容易混淆的OpenMP函数 获取线程的num，即ID。这里的ID是OpenMP的team内的ID，在OpenMP中，一个team内的线程的ID是俺顺序排列的，0、1、2... 说明：此函数在并行区域外或者并行区域内都可以调用。在并行区域外，获取的是master线程的ID，即为0。在并行区域内，每次执行到此函数，获取的是当前执行线程的ID。 此函数比较容易理解

多核编程 简单来说，由于现在电脑CPU一般都有两个核，4核与8核的CPU也逐渐走入了寻常百姓家，传统的单线程编程方式难以发挥多核CPU的强大功能，于是多核编程应运而生。按照我的理解，多核编程可以认为是对多线程编程做了一定程度的抽象，提供一些简单的API，使得用户不必花费太多精力来了解多线程的底层知识，从而提高编程效率。这两天关注的多核编程的工具包括OpenMP和TBB。按照目前网上的讨论，TBB风头要盖过OpenMP，比如OpenCV过去是使用OpenMP的，但从2.3版本开始抛弃OpenMP，转向TBB。 Linux下TBB安装 1） 到官方网站下载最新的TBB源程序。 https://www.threadingbuildingblocks.org/ 2） 建立安装目录，这个目录就是用来装TBB库的，当然我们得先在目录里编译TBB源程序。比如： /opt/tbb/ 3） 将下载的文件解压到2.2文件夹中，切换到2.2文件夹: cd /opt/tbb/ make make之后等一会就编译好了。 4） 在shell中运行： source /opt/tbb/ build/linux_* _release/tbbvars.sh （ “ * ” 号部分根据不同的系统情况而不同 ） 5） 进入/opt/tbb//example目录随便找个例子make一下，如果通过了，说明安装成功。 补充：

I'm using the new random number generators in in C++11. Although there are varying opinions, from this thread it seems that the majority believe they are not thread safe. As a consequence, I would like to make a program, where each thread uses its own RNG. An example is given in the related discussion of how to accomplish this with OpenMP: #include <random> #include <iostream> #include <time.h> #include "omp.h" using namespace std; int main() { unsigned long long app = 0; { //mt19937_64 engine((omp_get_thread_num() + 1)); //USE FOR MULTITHREADING mt19937_64 engine; //USE FOR SINGLE THREAD

For OpenMP, when my code is using the functions in its API (for example, omp_get_thread_num()) without using its directives (such as those #pragma omp ...), why directly specifying libgomp.a to gcc instead of using -fopenmp doesn't work, such as gcc hello.c /usr/lib/gcc/i686-linux-gnu/4.4/libgomp.a -o hello Update: I just found that linking to libgomp.a does not work, but linking to libgomp.so works. Does it mean OpenMP can not be static linked? Why -fopenmp only works without specifying the library files gcc hello.c -fopenmp -o hello Update: In other words, when using -fopenmp, why explicit

Problem I am learning about HPC and code optimization. I attempt to replicate the results in Goto's seminal matrix multiplication paper ( http://www.cs.utexas.edu/users/pingali/CS378/2008sp/papers/gotoPaper.pdf ). Despite my best efforts, I cannot get over ~50% maximum theoretical CPU performance. Background See related issues here ( Optimized 2x2 matrix multiplication: Slow assembly versus fast SIMD ), including info about my hardware What I have attempted This related paper ( http://www.cs.utexas.edu/users/flame/pubs/blis3_ipdps14.pdf ) has a good description of Goto's algorithmic structure.

I have a dual socket Xeon E5522 2.26GHZ machine (with hyperthreading disabled) running ubuntu server on linux kernel 3.0 supporting NUMA. The architecture layout is 4 physical cores per socket. An OpenMP application runs in this machine and i have the following questions: Does an OpenMP program take advantage (i.e a thread and its private data are kept on a numa node along the execution) automatically when running on a NUMA machine + aware kernel?. If not, what can be done? what about NUMA and per thread private C++ STL data structures ? The current OpenMP standard defines a boolean

I'm studying OpenMP's scheduling and specifically the different types. I understand the general behavior of each type, but clarification would be helpful regarding when to choose between dynamic and guided scheduling. Intel's docs describe dynamic scheduling: Use the internal work queue to give a chunk-sized block of loop iterations to each thread. When a thread is finished, it retrieves the next block of loop iterations from the top of the work queue. By default, the chunk size is 1. Be careful when using this scheduling type because of the extra overhead involved. It also describes guided

I have been developing an application to count circular objects such as bacterial colonies from pictures. What make it easy is the fact that the objects are generally well distinct from the background. However, few difficulties make the analysis tricky: The background will present gradual as well as rapid intensity change. In the edges of the container, the object will be elliptic rather than circular. The edges of the objects are sometimes rather fuzzy. The objects will cluster. The object can be very small (6px of diameter) Ultimately, the algorithms will be used (via GUI) by people that do