1.The comparison experiments for software pipeline and super block scheduling are done, which shows that the optimization effect can be increased by more than 30% and has obvious relation to instruction level parallelism (ILP).
对编译后端的软件流水和超块调度两种性能优化策略进行对比实验,表明其优化效果在30%以上,并且代码的指令级并行性(Instruction Level Parallelism,ILP)与优化效果存在明显的相关性.
2.ON PARALLELISM
浅论排比（PARALLELISM）手法
3.ILP (Instruction-Level Parallelism) is a technologyto enhance the processor performance by increasing the number ofinstructions executed per clock cycle.
指令级并行处理ILP（Instruction-Level Parallelism）是一项增强处理器性能的技术，它通过增加每个时钟周期执行的指令条数而提高性能。
4.The parallelism research for information retrieval involves data paralles lism and function parallelism. The data parallelism may show data-level par-allelisrm of SIMD system and data-set parallelism of distributed system or of MIMD system.
信息检索的并行性研究包括数据并行和功能并行，而数据并行可表现为SIMD系统中的数据级并行以及分布式系统或MIMD系统中的数据集并行。
5.The RAP/LOP-WAM parallel inference machine is based on the OR-forest description and is an extension of the WAM. It supports Restricted AND-Parallelism and Limited OR-Parallelism of logic programs.
RAP/LOP-WAM并行推理机基于OR树林描述,扩展WAM,开展逻辑程序的限制与/或并行性。
6.Image processing algorithms are inherently sub-word parallel(SWP),whose granularity is between data-level parallelism(DLP) and instruction-level parallelism(ILP).
图像处理算法的特点决定其是内在可并行的,这种并行粒度介于数据并行(DLP)和指令级并行(ILP)之间,称之为子字并行。
7.Brackets shall be spaced at proper intervals to provide rigidity and parallelism.
平台之间应该有适当的空间间隔，这样有利于提高坚硬性和平行性。
8.The concepts of regions in the Instruction Level Parallelism(ILP) and the Explicitly Parallel Instruction Computing(EPIC)are explained. Sevral algorithms of regional scheduling are introduced.
介绍指令级并行性 (IL P)中和指令级计算 (EPIC)中区域的直观概念和这些概念的形式化工作 ,并简单介绍区域的几个调度算法 ,为并行编译中的区域调度问题提供一个系统的、形式化的论述。
9.Based on the analysis and translation framework, an OpenMP compiling and optimization system is implemented on IA64/Linux, to provide a research vehicle for researches on related high performance computing and thread level parallelism exploration, and to be part of a larger OpenMP develop environment.
基于上面的分析和翻译框架,本文实现了一个IA64/Linux上的OpenMP编译与优化系统,以作为研究相关平台上高性能计算和开发线程级并行性的研究平台,同时也作为一个大的OpenMP开发环境的一部分。
10.On the Parallelism Analysis and Optimization of the Open Divx Video Encoder
Open Divx视频编码程序的并行性分析与优化研究
