Memory-Hierarchies-Matrix-Multiplication

A parallel computer is about how things are connected together the network that connects the parallel computers together and they can be connected in a number of different ways we’re going to tak...

1. Why Do We Need Parallel Programming 1.1 Why-Every Computer Is a Parallel Computer Nowadays, all computers are essentially parallel. The parallelism is found on all levels of a modern computer’s ...

可以并行化执行是因为CPU在设计时，增加了一些专用的向量寄存器，这些寄存器的长度往往大于通用寄存器，比如SEE的XMM寄存器，位宽为128位；AVX和AVX2的YMM寄存器，位宽为256位；AVX512的ZMM寄存器，位宽为512位。这些专用的向量寄存器可以同时放入多个数据。 变量定义 第一部分，统一为__m； 第二部分为位数如64、128、256等； 第三部位为变量类型，i表示in...