Integraing data parallel and task parallel can solve these problems.
数据并行与任务并行的结合可以很好地解决这些问题。
Accelerator is quite well suited for writing stencil-style data parallel programs.
Accelerator非常适合编写模板风格的数据并行处理程序。
Graphics CARDS are currently the best data parallel processing engines we have available.
显卡是目前最好的数据并行处理引擎，我们可用。
On the basis of the above, a data parallel algorithm of convolution calculation is proposed.
在此基础上给出了数据并行卷积计算算法；
That will end looking looking something like hadoop, mogilefs or S3 - a data parallel architecture.
那样最终就是像Hadoop、Mogilefs或S3一样的东西——并行数据架构。
PLINQ is a declarative model for performing data parallel queries against XML and in-memory objects.
PLINQ是一个对XML和内存对象进行并行查询的声明式的模型。
To select suitable grid resource for a data parallel application, an idea of pseudo-cluster is introduced.
为给数据并行应用选择适合的网格资源，文章给出一个伪机群的想法。
Or we could go a different direction and work on building an industry standard virtual ISA for data parallel architectures.
或者我们可以去不同的方向和工作建立一个行业标准的虚拟IS A的数据并行体系结构。
Being an important ramification of parallel processing, data parallel is widely used in scientific and technical computation.
作为并行处理的重要分枝，数据并行被广泛地应用于科学和工程计算中。
This is all in addition to writing kernels (data parallel functions) and actually using them in a program that does useful work.
这是所有除了书面仁(数据并行功能)，实际使用它们的计划，并有益的工作。
Then, the data parallel implementation scheme of zero-order interpolation and first-order interpolation of backward mapping are discussed.
几何操作反向映射的零阶内插与一阶内插的数据并行实现方法；
DPHL is a kind of data parallel high level modeling language, with which users can describe a problem solving process on an algorithmic level.
DPHL语言是一种数据并行高层建模语言，用于在算法层次上描述应用问题求解步骤。
With the increasing growth of data parallel applications, divisible job scheduling problem has been on the new focus of parallel scheduling area.
随着数据并行应用需求的日益增长，可任意划分负载的调度问题已经成为并行调度领域新的研究热点。
For implementing the data parallel computing (DPC) in grid that composed of multi-clusters, a grid development model based on multiagent is discussed.
在由多计算机机群构成的网格环境下，为了实现数据并行型计算，提出了一个基于多智能体机制的网格开发模型。
Firstly, this paper introduces two existed parallel execution models: SPMD and MPMD supporting data parallel and task parallel languages, respectively.
本文首先介绍两种典型的并行执行方式：支持数据并行语言的SPMD方式和支持任务并行语言的MPMD方式。
Even if you know the best and fastest algorithm for solving a data parallel problem, it isn't always possible to translate that to an efficient program.
即使你知道最好最快的算法，解决了数据并行问题，但是它并不总是能够把这种以高效率的计划。
As for data parallel problem, the author presents the TTPN model of a segment and the TTPN model involving several segments in logical process about parallel program.
对于数据并行问题，提出了并行程序逻辑进程中一个块的TTPN模型以及包含多个块的循环的TTPN模型。
This paper discusses the data parallel algorithm and its implementation on region growing image segmentation, the correctness is proved, and the performance is also analyzed.
针对基于区域增长的图像分割方法，讨论和研究了其数据并行实现方法，证明了这种数据并行方法的正确性，并简要对算法性能进行了分析。
But workload changes constantly in the network of workstations (NOWs), therefore load balancing is an important factor to influence the efficiency of data parallel application.
但是在工作站网络环境中，负载波动很大，负载平衡是影响其效率的一个重要因素。
The parallel and redundancy are implemented by using data communication method in emergency shutdown for enhancing the reliability of the system.
急停系统采用数据通信的方式实现并行冗余，提高了系统的可靠性。
The calculations are performed in parallel on each partition — where the data is.
计算是在每个分区(数据所在的位置)上并行执行的。
Service aggregation can be achieved using parallel process paths and data mapping.
可以使用并行流程路径和数据映射完成服务集合。
Performance can also be enhanced by striping the data over multiple data servers for parallel access.
还可以通过在多个数据服务器上分段数据进行并行访问来增强性能。
Then take advantage of any parallel structure in the data to effectively add more capacity using clusters.
然后，利用数据中的任何并行结构，通过聚集有效地增加容量。
His recent focus has been on high performance computing including in-memory data grid, parallel and grid computing.
他最近的关注点是高性能计算，包括内存数据网格、并行计算和网格计算。
Also, the response time of complex analytical queries can be reduced by spreading the data across multiple database partitions, such that all partitions work on their assigned data in parallel.
此外，通过跨多个数据库分区分散数据以使多个分区可并行处理分配到其上的数据，也能减少复杂的分析查询的响应时间。
In the same way, the generated intermediate data is processed in parallel, making the approach ideal for processing very large amounts of data.
使用相同的方法，已生成的中间数据将被并行处理，这是处理大量数据的理想方法。
Microsoft Research's Accelerator Project exposes a.net library for performing parallel data processing using a computer's GPU.
微软研究院加速器项目公开了一个使用计算机的GPU完成并行数据处理的。NET函数库。
Figure 21 shows how you can set up an aggregation of data for parallel ownership scenarios.
图21演示如何为并行所有权场景设置数据聚合。
Figure 21 shows how you can set up an aggregation of data for parallel ownership scenarios.
图21演示如何为并行所有权场景设置数据聚合。
应用推荐