基于热例程的动态二进制翻译优化

依据对系统级程序行为特性的观察,提出了一种基于热例程的动态二进制翻译优化方法。该方法以频繁执行的例程作为优化单位,通过块内和块间优化算法消除动态二进制翻译引入的冗余。相比基于踪迹的优化方法,该方法具有优化单位发现开销更小、代码区域更大、无重复翻译等优点,更适用于系统虚拟机中操作系统代码的优化。在跨平台系统虚拟机监控器ARCH-BRIDGE上的测试表明,通过对内核代码实施该优化方法,SPEC CPUINT 2006程序的效率提升了3.5%~14.4%,相比基于踪迹的优化,性能最大提升了5.1%。     

如何编写有利于编译器优化的代码?

引言对于嵌入式系统,最终代码的体积和效率取决于由编译器生成的可执行代码,而非开发人员编写的源代码;但是,源代码的优化可以帮助编译器生成更加优质的可执行代码。因此,开发人员不仅要从整体效率等因素上去构思源代码体系,也要高度关注编译器的性能和编译优化的便捷性。     

基于目标代码大小的指令选择

在传统编译器中,指令选择往往采用动态规划的方法,其目的是优化目标代码性能（减小程序运行时间）。在嵌入式系统中,受到成本的约束,一般只有很有限的存储空间,因此要求目标代码仅可能地小。本文针对具有可变长指令的处理器,以优化代码大小为目的,采用动态规划的方法进行指令选择;相对于优化性能的目标代码,可缩小代码大小的15%
到20%。     

VLPL-S在Knights Landing上的优化与性能评估

VLPL-S代码是基于Particle-in-Cell(PIC)算法开发的激光等离子体模拟程序,PIC算法是激光等离子模拟领域的常用主流算法之一。讲述了VLPL-S代码在Intel?新推出的Knights Landing平台上的早期移植及优化工作。通过采用在代码优化中常用的优化方法,例如访存优化、多线程优化、向量化,为VLPL-S代码实现了1.68倍的加速比。对于优化以后的VLPL-S代码,其在Knights Landing 7210P单节点上的性能是其在双路Xeon E5-2697v4节点上性能的1.53倍。还对比了不同优化方法在Knights Landing及Xeon平台上所获得的性能提升。结果表明,对于VLPL-S代码,以往CPU代码优化工作中常用的优化方法在新的Knights Landing平台中同样有效。     

优化代码与充分利用系统资源

本文通过比较运行时间和代码体积，以标准C语言为例，从几个方面着手具体描述了如何优化代码，实现系统资源的充分利用。     

基于即时编译的动态污点跟踪优化

动态污点跟踪技术展现了在移动隐私保护方面强大功能,但存在系统性能较低问题.提出了一个基于即时编译的动态污点传播优化方法.首先,将程序逻辑精确抽象为污点传播逻辑,简化污点传播分析复杂性;然后,提出了一个污点传播框架并证明了在该框架下污点传播分析的正确性和有效性;最后,采用消除,替换和移动等方法将冗余低效的污点传播代码转化为高效等价的污点传播代码.实验结果表明经过优化后单条热路径的污点传播代码节省了38%内存占用和指令执行时间,系统整体性能平均提升了6.8%.     

浅议Java程序优化的几种方法与成效

本文介绍了Java代码优化的过程,总结了优化Java程序的一些最佳实践,分析了进行优化的方法并解释了性能提升的原因。从多个角度分析导致性能低的原因并逐个进行优化使得程序性能得到极大提升,代码可读性、可扩展性更强。、     

一种基于SLP的新型编译框架

对于SLP算法不能高效处理并行代码占有率较小的大型应用程序的问题,本文提出并评估了一种新型的基于改进的SLP(Superword level parallel)算法的编译框架。它主要包括三个阶段,首先,将代码中的结构相似的异构语句通过改进的SLP算法尽可能的改为同构语句;然后,用全局的观点,在优化目标代码之前获取其数据模型重用;最后,联合数据布局优化进行进一步的性能提升。本文就此框架做了大量实验,实验结果表明本框架比SLP算法性能更佳,优于它约15.3%。     

CYBER205上的向量优化

向量优化是指在一台给定机■上,对一个已知的向量计算生成最好的结果代码。本文为在控制政据 CYBER 205上运行的标量和向量源代码提出了性能分析模型。这些模型的精度对于标量代码一般是在30%以内,对于向量代码一般是在10%以内。对于一个已知的并行计算,如果编译程序可以产生一个以上的代码程序,则可以利用来自这些模型的性能估计,以便选择应当执行哪一个代码程序。把含有两个或更多个源代码版本的16个FORTRAN 核作为测试这个方法的标准检查程序。13个核正确地进行了“向量优化”。没有进行常规优化的3个核有17%的平均性能损失。用一批核作为标准检查程序表明向量优化把它的性能改进了四倍多,所有已经正确优化了的核,有98%的在程序上也可改进。     

浅析嵌入式系统编程中的代码优化

随着嵌入式系统应用复杂度的提高,高效精简的软件代码是提高系统性能的关键因素之一。文章首先分析了嵌入式编程的特点,阐述了内联函数、循环展开、软件流水等几种代码优化方法,最后提出了应用行程编码(RLE)方法来减少常量数组对存储空间的占用。实验表明,经上述方法优化后的目标代码,执行效率提升了近50%,有效地提高了系统性能。     

Map Reduce inspired loop mapping for coarse-grained reconfigurable architecture

Our work investigates how to map loops efficiently onto Coarse-Grained Reconfigurable Architecture （CGRA）. This paper examines the properties of CGRA and builds MapReduce inspired models for the loop parallelization problem. The proposed model has a more detailed performance metric and a more flexible unrolling scheme that can unroll different loop levels with different factors. A Geometric Programming based approach is proposed to resolve the optimization problem of loop parallelization problem. The proposed approach can find the optimal unrolling factor for each level loop, resulting in better parallelization of loops. Experimental results show that the proposed approach achieved up to 44% performance gain compared to the state-of-the-art loop mapping scheme.     

基于多层次并行架构的H.264解码算法

针对高清图像视频的实时解码需求,提出一种基于多层次并行流水架构的解码算法。该算法首先针对图像的宏块行实现基于功能模块的行级并行算法,并通过功能模块的二次划分进行核间负载均衡的优化,再针对解码过程中开销较大的滤波环节,利用宏块之间的依赖关系进行多核并行处理,对行级并行算法进行更深层次上的再优化设计。实验结果表明,该算法可以在TILEPro64平台上实现1080P全高清码流的实时解码,实现了较高的并行加速比,最高达到10.01,和已有的并行解码算法相比,加速比提升80%。     

嵌入式处理器片外访存加密机制设计与实现

高安全敏感领域的嵌入式系统面临总线监听、数据篡改、离线分析等类型的恶意攻击,试图窃取密码、篡改信息等。特别是配合硬件电路的攻击,给用户造成重大的损失。为了从根本上解决系统外部电路系统攻击威胁,提出片外访存加密认证机制,选择AES-GCM算法,对所有片外写数据进行加密,对读数据进行解密并认证。同时设计一次密码与页地址置乱函数产生二次密钥,保障了加密强度。进一步通过软件实现LRU Cache优化性能,在STM32系列微处理器硬件平台上,软件实现片外访存加密认证机制。在内存压力测试中,加密片外访存性能平均降低了9%。     

基于量子粒子群优化算法的压缩感知数据重构方法

针对传感器监测对象特点,将压缩感知理论应用于数据压缩过程以降低通信能耗,并根据现有压缩感知数据重构算法存在的重构精度受稀疏度影响较大的缺点,在分析了压缩感知数据重构原理后,提出了将原始信号按固定长度进行分帧处理以减少算法解空间的数量,并将量子理论中的编码方式应用于粒子群优化算法,提出了基于量子粒子群优化算法的压缩感知数据重构方法QP-CSDR。算法根据传感器监测对象特点,从统计学角度出发对粒子群优化算法中的粒子初始位置及粒子群更新方式加以改进,以提高数据重构精度。仿真实验结果表明,在稀疏度小于50的条件下,QP-CSDR算法相对已有算法在重构精度方面性能提升20%~40%,该算法已应用于微地震及音频监测系统中,经实际检验算法在保证数据精度的前提下延长系统寿命2倍~4倍左右。     

飞翼式飞行器结构布局与构件尺寸的两级优化

为优化飞翼式飞行器的结构,提出同时考虑结构布局优化和构件尺寸优化的两级优化方法. 第1级优化将翼梁数量范围和位置范围作为约束,以重量最轻为优化目标,采用iSIGHT的多岛遗传算法优化结构布局;第2级优化给予第1级给定的结构布局方案,在满足应力约束和位移约束的前提下,确定各构件最佳尺寸,使该结构布局方案的结构重量最轻. 在第2级优化中,根据CAD外形模型和固定的结构布局参数,采用MSC Patran的PCL语言自动生成飞机的结构有限元模型,采用MSC Nastran优化全机结构,并将优化结果返回给第1级. 再通过两级之间的迭代获得结构布局和构件尺寸的最优设计方案. 整个两级优化过程用iSIGHT集成. 实例表明该方法为飞翼式飞行器结构布局和构件尺寸协同优化问题提供1种有效的解决办法.     

Combined size and shape optimization of structures with a new meta-heuristic algorithm

In this study, a new meta-heuristic algorithm called teaching-learning-based optimization (TLBO) is used for the size and shape optimization of structures. The TLBO algorithm is based on the effect of the influence of a teacher on the output of learners in a class. The cross-sectional areas of the bar element and the nodal coordinates of the structural system are the design variables for size and shape optimization, respectively. Displacement, allowable stress and the Euler buckling stress are taken as the constraint for the problem considered. Some truss structures are designed by using this new algorithm to show the efficiency of the TLBO algorithm. The results obtained from this study are compared with those reported in the literature. It is concluded that the TLBO algorithm presented in this study can be effectively used in combined size and shape optimization of the structures.     

PSO_BFA优化词袋模型及蛋白质亚细胞定位预测

提出了一种基于PSO_BFA优化的词袋模型。传统词袋模型有两个重要参数：窗口大小[d]和字典大小[k]。结合粒子群算法和细菌觅食算法产生新的PSO_BFA混合优化算法,在PSO进行局部搜索时,加入BFA的复制和迁移行为,得到PSO_BFA的最优解即为窗口大小和字典大小的最佳组合。将优化词袋模型与蛋白质序列的氨基酸组成算法和伪氨基酸组成算法结合,获得蛋白质序列的词袋特征。实验结果证明,基于PSO_BFA优化的词袋模型能有效提高蛋白质亚细胞定位预测的精度。     

Two coding based adaptive parallel co-genetic algorithm with double agents structure

This paper systematically proposed a multi-population agent co-genetic algorithm with double chain-like agent structure (MPATCGA) to solve the problem of the low optimization precision and long optimization time of simple genetic algorithm in terms of two coding strategy. This algorithm adopted multi-population parallel searching mode, close chain-like agent structure, cycle chain-like agent structure, dynamic neighborhood competition, and improved crossover strategy to realize parallel optimization, and has the characteristics of high optimization precision and short optimization time. Besides, the size of each sub-population is adaptive. The characteristic is very competitive when dealing with imbalanced workload. In order to verify the optimization precision of this algorithm with binary coding, some popular benchmark test functions were used for comparing this algorithm and a popular agent genetic algorithm (MAGA). The experimental results show that MPATCGA has higher optimization precision and shorter optimization time than MAGA. Besides, in order to show the optimization performance of MPATCGA with real coding, the authors used it for feature selection problems as optimization algorithm and compared it with some other well-known GAs. The experimental results show that MPATCGA has higher optimization precision (feature selection precision). In order to show the performance of the adaptability of size of sub-populations, MPATCGA with sub-populations with same size and MPATCGA with sub-populations with different size are compared. The experimental results show that when the workload on different sub-populations becomes not same, the adaptability will adaptively change the size of different sub-population to obtain precision as high as possible.     

嵌入式系统程序优化方法的研究

嵌入式系统往往对实时性、系统功耗和程序代码长度有特殊的要求，本文从程序设计的角度讨论满足这些要求的程序代码优化方法。文中首先讨论了程序执行时间的优化方法，随后讨论了程序代码长度的优化方法，最后讨论了程序功耗的优化方法。     

Optimization design of general triglide parallel manipulators

Optimization design of parallel manipulators has attracted much interest from researchers in recent years. The reported methodologies attempted to achieve optimal design of parallel manipulators considering several properties, such as dexterity, stiffness, and space utilization, which are important parameters to be considered. However, stiffness analysis considered by many researchers generally ignores the deformation of the mobile platform. For space utilization, there is no reported method to consider the variation in the physical size caused by different postures of the manipulator. Additionally, although optimization of a linear delta and an orthoglide has been presented by several researchers, optimization of a general triglide has not been reported. In order to address these issues, this paper presents a multi-objective optimization addressing dexterity, stiffness, and space utilization of a general triglide. Its stiffness matrix is obtained considering the deformation of mobile platform, limbs, and actuators. A novel stiffness index is used to evaluate its stiffness property considering external wrench applied on the manipulator. The physical size of the triglide is represented using both a constant size and a variable size. Comparing with a reported optimization methodology, it is proven that the proposed method is capable of providing optimal solutions with better properties.