串行程序

一种将串行程序划分成并行任务的方法

A Method for Partitioning a Sequential Program into Parallel Tasks

该算法通过识别并行循环任务以及通过确定调用上下文关系识别并行过程任务，来划分串行程序中的并行成份。该算法是对串行程序实施并行分解的基础。

This algorithm is the basis for decomposing parallelism in a sequential program and is implemented with recognizing parallel loop tasks and parallel procedural tasks by determining the call context .

用Petri网分析串行程序的并行性

Apply Petri net to analyse the parallelism of serial program

首先对传统的串行程序进行分析，并利用Intel性能分析器找到串行程序的热点代码。

The traditional serial-execution procedures were analyzed , and the Intel performance analyzer was used to find the hot spot in the serial-execution program .

因此，如何合理地对串行程序进行分析，评估串行程序并行化后在GPU上的执行效率变得尤为重要。

Therefore , how to analyse the serial program reasonably and to predict the efficiency of parallel program on GPU becomes particularly important .

这给传统的串行程序设计带来了严峻的挑战，在计算行业中，已经逐步的发展成为CPU与GPU共同合作计算的协同处理的时代，而并非传统的只使用CPU进行计算的中央处理。

Traditional serial programming is facing serious challenges . The computer industry has changed from the central processing of CPUs to the cooperative processing of both CPUs and GPUs .

通过该模型可以预测出串行程序用GPU加速时的时间开销，将其与串行执行的开销进行对比，从而判断是否用于GPU加速，进而指导串行程序的并行化。

The model can estimate the total time cost of parallel program on GPU , which can determine whether it is worthy for GPU acceleration .

因此，本文利用二进制翻译技术并结合CPU和GPU新型异构体系结构来提高传统串行程序的执行性能。

Therefore , this paper uses binary translation technology and the new heterogeneous architecture of CPU and GPU to improve the performance of many traditional serial programs running on CPU .

其主要工作有：1对ABEEMσπ力场模型中能量求解的串行程序代码进行分析。

The main work : 1 ABEEM σπ force field of energy on solving the serial code for analysis .

首先对串行程序进行分段，提出并行划分模型，基于这个模型提出一种并行划分算法PDMA；

The serial program is divided into segments and the definition of parallelism model is brought forward . A parallelism algorithm named PDMA is given based on this model .

但是，随着分子体系规模的增大，利用ABEEMσπ模型电荷分布串行程序计算电荷分布的时间越来越长。

However , with the increasing of the size of molecular system , the time of calculating the charge distribution by the serial program of ABEEM σπ model gets longer and longer .

通过矩阵乘法实验表明，静态二进制翻译系统SBTG翻译生成的并行程序不仅能正确执行传统串行程序的功能，而且执行速度比传统串行程序要快。

Through the matrix multiplication results show that the parallel program which is generated by the static binary translation system SBTG not only properly perform the function of traditional serial program , but also its execution speed faster than the implementation of traditional serial procedures .

但并行程序设计与串行程序设计相比较在实现上要更复杂。

Parallel programming is more complex compared with their sequential counterparts .

串行程序并行化技术研究与一种新实现构想

Research of Serial Program Parallelization Technology and a New Design Idea

并行化编译器通过发掘串行程序中的并行性来提高程序的运行性能。

Parallelization compiler improves performance by exploiting the parallelism in serial program .

本文探讨并行预编译系统的设计，并着重讨论串行程序中的并行成分识别算法。

The design of a parallel pre-compiler is discussed in this paper .

其中计算单元是传统串行程序模块。

The computing units are classic serial program modules .

对运算结果进行了分析并给出了与串行程序相比较的并行加速比。

The result is analyzed and the actual paralleling speed ratio is given .

串行程序在大粒度级的并行分解及可并行执行包的形成

Decomposing parallelism in a sequential program at large-grained and forming parallel executable packages

提出了将串行程序并行化的方法。

The solution of serial program parallelism is presented .

与串行程序相比，并行程序调试会遇到新的问题。

Compared with serial programs , parallel programs debugging will face some new problems .

而绝大多数目前已经开发的实时系统测试和调试工具是针对串行程序的（非实时）。

However , most debugging technique is for non real-time systems and sequential program .

并行化编译器可以把现有的串行程序自动或半自动地转换为并行程序。

Parallelizing compiler can transform serial programs to parallel programs automatically or semi automatically .

串行程序的依赖关系分析和向量化

Dependence analysis and vectorization for serial programs

分析讨论了分子动力学模拟的算法特征和计算特点，对串行程序作了优化，并使之适合于作并行化。

Analyzes the algorithmic character of molecular dynamics simulation , and optimizes the serial program .

基于处理器上多个核的编程环境，给出了串行程序并行化的设计流程。

And design a procedure for making the serial process software running parallel on the multi-core processors .

一些传统串行程序测试方法和理论不再适用于并发程序的测试。

Some of the traditional testing methods and theories of sequential programs are not applicable in concurrent program testing .

分别采用聚合通信技术和非阻塞通信技术对原串行程序进行了并行化处理，并编制相应代码。

Process the serial program in parallel with polymerization communication technology and non-blocking communication technology , and edit relevant codes .

如果单独考虑，那么进程仅仅是一个独立的串行程序，它只与外部I/O设备交互。

Considered individually , a process is just an independent serial program that interacts only with external I / O devices .

最后将并行程序与串行程序的计算结果相比较，验证了该并行程序的正确性。

The computation result is identical to that of series program , and the correctness of the parallel program is verified .