Фильтры

Блоги

最快线程间数据交换算法,有效避免锁竞争 -- TwoQueues

处理多线程数据共享问题注意的几个要点:

1、锁竞争:尽量减少锁竞争的时间和次数。

2、内存:尽量是使用已分配内存,减少内存分配和释放的次数。尽量是用连续内存,减少共享占用的内存量。

多线程数据交换简单方案A:

定义一个list,再所有操作list的地方进行加锁和解锁。

简单模拟代码:

Автор: Последнее обновление: 04.07.2019 - 21:30
Article

Vectorizing Loops with Calls to User-Defined External Functions

Introduction

Автор: Anoop M. (Intel) Последнее обновление: 12.12.2018 - 18:00
Блоги

The switch() statement isn't really evil, right?

In my current position, I work to optimize and parallelize codes that deal with genomic data, e.g., DNA, RNA, proteins, etc.

Автор: Clay B. (Blackbelt) Последнее обновление: 04.07.2019 - 10:46
Article

Improve Intel® MKL Performance for Small Problems: The Use of MKL_DIRECT_CALL

One of the big new features introduced in the Intel® Math Kernel Library (Intel® MKL) 11.2 is the greatly improved performance for small problem sizes.

Автор: Zhang, Zhang (Intel) Последнее обновление: 07.07.2019 - 10:35
Блоги

opencl_node overview

Introduction
Автор: Alex (Intel) Последнее обновление: 30.05.2018 - 07:08
Article

使用 OpenCL™ 2.0 读写图片

While Image convolution is not as effective with the new Read-Write images functionality, any image processing technique that needs be done in place may benefit from the Read-Write images. One example of a process that could be used effectively is image composition. In OpenCL 1.2 and earlier, images were qualified with the “__read_only” and __write_only” qualifiers. In the OpenCL 2.0, images can...
Автор: Последнее обновление: 31.05.2019 - 14:20
Article

Using Intel® MPI Library 5.1 on Microsoft* Windows* with Microsoft* MPI based applications

Why it is needed?
Автор: Dmitry S. (Intel) Последнее обновление: 12.12.2018 - 20:11
Блоги

Reduce Boilerplate Code in Parallelized Loops with C++11 Lambda Expressions

Parallelize loops with Intel® Threading Building Blocks using Intel® C++ Compiler for lambda expressions.
Автор: gaston-hillar (Blackbelt) Последнее обновление: 12.12.2018 - 18:00
Блоги

Debug Intel® Transactional Synchronization Extensions

If printf or fprintf functions cause transaction aborts, use Intel® Processor Trace as a work-around.
Автор: Roman Dementiev (Intel) Последнее обновление: 04.07.2019 - 17:00
Article

Implementing a Masked SVML-like Function Explicitly in User-Defined Way

The Intel® Compiler provides SIMD intrinsics APIs for short vector math library (SVML) and starting with Intel® Advanced Vector Extensions

Автор: Последнее обновление: 16.07.2019 - 08:37