On vector processors a register is treated both as a single unit and as multiple units.
在向量处理器上，寄存器可以作为一个单元进行处理，也可以作为多个单元进行处理。
Clusters are built using commodity hardware and cost a fraction of the vector processors.
集群都是使用常见的硬件进行构建的，其成本只是向量处理器的很小一部分。
This was especially obvious when we compared vector processors with scalar processors in Part 1.
在第1部分中对向量和标量处理器进行比较时，这尤其明显。
Special compilers for programming languages like FORTRAN were developed to optimize the code to take advantage of the vector processors.
开发了用于像FORTRAN的程序设计语言的特殊编译器来优化代码，以利用向量处理器。
SIMD optimization is very similar to traditional automatic vectorization for vector processors.
这种优化和传统的针对向量处理器的自动向量化非常类似。
SIMD optimization is very similar to traditional automatic vectorization for vector processors.
这种优化和传统的针对向量处理器的自动向量化非常类似。
应用推荐