In practice neural networks create a new kind of architectures of highly parallel computations.
人工神经网络在许多应用领域中建立了一种新的高度并行计算的结构。
After parallel algorithms came grid computing, which ran parallel computations on idle desktops.
并行算法发明以来，首先迎来的是网格计算，网格计算是借助空闲的桌面计算机资源进行并行计算。
But the question remains, how do all those parallel computations become integrated, and how does the self feel that "I" made the decision?
但问题依然存在，所有这些并行计算是如何整合起来的，自我如何感觉到是“我”做了这个决定？
MapReduce breaks down a problem into millions of parallel computations in the Map phase, producing as its output a stream of key-value pairs.
MapReduce在映射阶段将一个问题分解为数百万个并行计算，并生成键-值对流作为输出。
One approach to handling this broad set of problems efficiently is through massively parallel computations on relatively inexpensive hardware.
有效处理大量此类数据的一种方法是在相对比较廉价的硬件上执行大量并行计算。
For 3d elasticity problems, the parallel computations based on the fast multipole and the conventional boundary element method (BEM) on PC cluster are compared.
以三维弹性力学问题为例，对快速多极与常规边界元法机群并行计算进行了比较。
Different from traditional parallel computations algorithms, Space-Time Adaptive Processing (STAP) algorithms are composed of several phases and each phase consists of many identical, separable tasks.
与传统并行计算研究对象不同的是，空时自适应处理(STAP)的各阶段是由多个相同的、可分离的任务组成。
They "get" concurrency and many of the issues required to implement fast, accurate computations in parallel.
他们“获得”并行性，并且让许多实现快速、准确计算所需的问题并行处理。
If the same processing is required for each array element, with no dependencies in the computations, and no communication required between tasks, we have an ideal parallel computing opportunity.
如果每个数组的元素都需要被处理，而且数组间没有依赖关系，执行的计算任务之间也不需要通信，这样的话将是一个执行并行式计算的理想环境。
Instead of using specially optimized vector hardware, it USES standard scalar processors, but in large Numbers to perform several computations in parallel.
它不使用专门优化过的向量硬件，而是使用标准的标量处理器，但是它采用了大量的处理器来并行处理多个计算任务。
This distribution implies parallel computing since the same computations are performed on each CPU, but with a different dataset.
因为计算分布到不同的CPU，且每个CPU处理不同的数据集，所以这样的分布式处理意味着可以采用并行计算。
The process of calculating the result can occur in parallel with other computations.
对结果进行计算的过程能并行地发生在其他计算过程当中。
Cloud computing grew out of parallel computing, a concept that many problems can be solved faster by running the computations in parallel.
云计算诞生于并行计算，即多个问题可以通过同时在多台计算机上运行来加快速度。
But passing messages between cores is much more time-consuming than executing computations on a given core, and it ends up eating into the gains afforded by parallel execution.
但是在处理器间传递信息比在某一处理器上进行运算要费时得多，这样下来，进行平行运算的优势就没有了。
The computations are relatively straightforward, but they must be able to cope with all conceivable geometric situations, including vertical lines, parallel lines, and so on.
计算是相对地直接的，但他们必须能应付以所有可以想像的几何学情况，包括垂直线，平行的线，等等。
Matrices computation is the most important in numerical computations. Linear array with reconfigurable pipelined bus system (LARPBS) is a parallel efficient computational model based on optical bus.
矩阵运算是最重要的数值计算，基于流水光总线的可重构线性阵列系统(LARPBS)是一种建立在光总线上的并行高效计算模型。
Matrices computation is the most important in numerical computations. Linear array with reconfigurable pipelined bus system (LARPBS) is a parallel efficient computational model based on optical bus.
矩阵运算是最重要的数值计算，基于流水光总线的可重构线性阵列系统(LARPBS)是一种建立在光总线上的并行高效计算模型。
应用推荐