Prophet推测多线程系统设计与实现

推测多线程技术通过推测执行的方式开发应用程序的线程级并行性,以提高程序执行性能。该技术一般通过执行模型来检测运行时可能的线程推测错误情况,并采取合适的机制恢复程序正确运行。描述的Prophet是一种基于硬件实现的推测多线程执行模型。重点描述了Prophet执行模型针对执行模型设计的关键问题的解决方案,包括Prophet的线程状态控制和多版本的Cach。系统,Prophet的多版本Cache系统提供了推测数据缓存功能,并使用基于总线监听的Cache协议实现了数据依赖违规检测。还给出了使用Olden基准程序对Prophet执行模型进行功能和性能测试的结果,并分析说明了Prophet系统可以有效地开发应用程序的线程级并行性。     

线程切换开销分析工具的设计与实现

在多片多核计算机系统中,线程切换的间接开销受到体系结构、负载模式和调度策略的影响,为了获得更为稳定的测试结果,在分析Lmbench测试程序框架和进程切换测试原理的基础上,针对多片多核处理机系统中的线程测试需求,通过集成多种负载模式和调度策略,设计和实现了新的线程切换延时测试程序LTC,为多核系统下的线程切换延时测试与分析提供了有效手段。     

多核并行技术在分子动力学模拟中的应用

为了充分利用多核处理器资源,研究了一种用于分子动力学模拟中的多核并行技术。在多核处理器上利用OpenMP技术实现多线程创建与同步、动态设置子线程的调度运行方式以及负载均衡以减少子线程执行等待时间。通过对不同分子体系结构下的动力学模型测试,得出在不同子线程下并行计算的时间,并且得到了良好的性能加速比。实验结果表明,采用OpenMP并行技术可有效地提高电荷求解过程在分子动力学模拟运算中的时间效率,以及多核计算机资源的利用率。     

一种海量数据流应用并行优化模型

计算进入了多核时代,处理器的发展不再由更快的主频带动,而是依靠增加片上的多个核心.但是,对于高性能应用来说,多核平台的并行处理由于缺少适合的并行程序开发工具还处于初始阶段,对应用的优化需要对底层线程结构的深入了解和正确使用.从海量数据流应用的特点出发,提出了三级流水多线程模型,它的线程同步机制没有竞争,并且实现了不同特征数据流的差别服务.然后,在遥感图像处理和骨干网网络入侵检测系设计中,应用了海量数据流应用模型,并在多个多核平台下对骨干网网络入侵检测系统进行了性能评价.对SPARC T1平台上的线程映射方法进行研究,测试了不同映射方法的性能,并归纳出应用在体系结构方面的特征;采用Sun SPARC T1架构8核32线程服务器和曙光x86架构8处理器16核服务器对系统吞吐率进行了测试,实验结果都表现了良好的可扩展性;使用真实骨干网络流量记录文件回放产生的模拟流量,对比测试了模型应用前后数据流的服务时间,改进系统的响应时间获得了显著的提高;针对系统连接数大、负载重和处理多样性的特点,采用基于探针流的采样算法准确测试了在精确预测IP网段策略下系统的服务质量,同时也测试了增加服务质量优化后系统的延迟开销,实验结果表明,系统在引入较少延迟下提高了数据流的服务质量.     

编程型游戏的线程调度模型设计与实现

编程型游戏采用多线程技术实现对角色控制程序的执行和调度。阐述了编程型游戏的线程调度模型和基于.NET框架的实现技术,包括协作多线程调度模型、调度线程的算法、基于AutoResetEvent的线程同步技术、角色控制程序的基本结构和线程管理模块的构成。测试结果表明,采用所述线程调度模型和技术的线程管理模块能够对角色线程进行有序调度,性能良好。     

基于TCL的IPv6中BGP一致性测试系统设计

根据协议一致性测试的特点和要求,针对BGP4＋协议的特点,介绍了一种BGP4＋协议一致性测试系统的设计方法。该方法使用带有控制通道的多个虚拟测试器环境,用TCL语言作为测试例的描述语言,具有很好的测试能力和灵活性。在介绍BGP4＋的基础上,设计并实现了该协议一致性测试系统,并对一基于FreeBSD平台的BGP4＋一致性测试实现的结果进行了分析。     

面向多核处理器的多实例并行BGP协议模型设计与实现

开发BGP的线程级并行可以满足不断膨胀的互联网应用对协议性能的迫切需求。本文提出了一种面向多核处理器的多实例并行BGP协议模型,它以邻居会话划分为基础,借助数据并行思想实现了不同邻居会话在多个线程上的并行处理。首先,通过分析BGP协议特点,我们给出了该模型的总体框架设计,然后在Quagga BGP的基础上完成了具体实现。在Intel Xeon四核服务器上的性能测试结果显示,多实例并行BGP协议较BGP协议的性能加速均值在2.73左右,并且能够有效利用多核处理器计算资源,为改善协议处理能力提供了更大的提升空间。     

一种云计算环境下的虚拟机动态迁移策略

提出一种云计算环境下的虚拟机动态迁移策略DMS-VM(Dynamic Migration Strategy for Virtual Machine) 。首先,假设了一种云计算环境下虚拟机迁移场景,在该场景下多种应用服务请求处于动态变化之中,并且提供的应用服务与虚拟机是一对一绑定的;其次,提出了该场景下的基于多约束的多目标规划模型,并设计遗传算法作为虚拟机的主要迁移策略;最后利用某个企业的大数据中心作为云端测试环境, 对比测试DMS-VM算法与已有的虚拟机迁移算法的性能。实验结果表明,DMS-VM迁移策略能够更好的减少物理主机的使用数量和虚拟机的迁移次数,同时降低数据中心能耗,性能优于已有的迁移策略。     

区分服务网络中基于SLA的协商机制研究

针对区分服务网络中基于服务级别约定（SLA）的实时协商问题,在分析协商优化过程后,提出一种满足确保转发服务的自动协商描述模型,模型包括一对一和一对多两种协商场景。通过模拟验证了模型的收敛性,并分析了在不同服务质量值和价格调节值下达成的协议方案性能。     

IPv6中IS-ISv6协议的一致性测试系统设计与实现

根据协议一致性测试的特点和要求,针对IS-IS协议的特点,介绍了一种IS-ISv6协议一致性测试系统的设计方法.该方法使用带有控制通道的多个虚拟测试器环境,用Tcl语言作为测试例的描述语言,具有很好的测试能力和灵活性.本文在介绍IS-ISv6的基础上,设计并实现了IS-ISv6的协议一致性测试系统,并对一基于FreeBSD平台的IS-ISv6一致性测试实现的结果进行了分析.     

Top-down induction of model trees with regression and splitting nodes

Model trees are an extension of regression trees that associate leaves with multiple regression models. In this paper, a method for the data-driven construction of model trees is presented, namely, the stepwise model tree induction (SMOTI) method. Its main characteristic is the induction of trees with two types of nodes: regression nodes, which perform only straight-line regression, and splitting nodes, which partition the feature space. The multiple linear model associated with each leaf is then built stepwise by combining straight-line regressions reported along the path from the root to the leaf. In this way, internal regression nodes contribute to the definition of multiple models and have a "global" effect, while straight-line regressions at leaves have only "local" effects. Experimental results on artificially generated data sets show that SMOTI outperforms two model tree induction systems, M5' and RETIS, in accuracy. Results on benchmark data sets used for studies on both regression and model trees show that SMOTI performs better than RETIS in accuracy, while it is not possible to draw statistically significant conclusions on the comparison with M5'. Model trees induced by SMOTI are generally simple and easily interpretable and their analysis often reveals interesting patterns.     

Daily prediction of short-term trends or crude oil prices using neural networks exploiting multimarket dynamics

This paper documents a systematic investigation on the predictability of short-term trends of crude oil prices on a daily basis. In stark contrast with longer-term predictions of crude oil prices, short-term prediction with time horizons of 1-3 days posits an important problem that is quite different from what has been studied in the literature. The problem of such short-term predicability is tackled through two aspects. The first is to examine the existence of linear or nonlinear dynamic processes in crude oil prices. This sub-problem is addressed with statistical analysis involving the Brock-Dechert-Scheinkman test for nonlinearity. The second aspect is to test the capability of artificial neural networks (ANN) for modeling the implicit nonlinearity for prediction. Four experimental models are designed and tested with historical data: (1) using only the lagged returns of filtered crude oil prices as input to predict the returns of the next days; this is used as the benchmark, (2) using only the information set of filtered crude oil futures price as input, (3) combining the inputs from the benchmark and second models, and (4) combing the inputs from the benchmark model and the intermarket information. In order to filter out the noise in the original price data, the moving averages of prices are used for all the experiments. The results provided sufficient evidence to the predictability of crude oil prices using ANN with an out-of-sample hit rate of 80%, 70%, and 61% for each of the next three days’ trends.     

From sequential programs to concurrent threads

Chip multiprocessors are of increasing importance due to difficulties in achieving higher clock frequencies in uniprocessors, but their success depends on finding useful work for the processor cores. This paper addresses this challenge by presenting a simple compiler approach that extracts non-speculative thread-level parallelism from sequential codes. We present initial results from this technique targeting a validated dual-core processor model, achieving speedups ranging from 9-48% with an average of 25% for important benchmark loops over their single-threaded versions. We also identify important next steps found during our pursuit of higher degrees of automatic threading.     

A genetic method for designing TSK models based on objective weighting: application to classification problems.

This paper proposes a genetic-based algorithm for generating simple and well-defined Takagi-Sugeno-Kang (TSK) models. The method handles several attributes simultaneously, such as the input partition, feature selection and estimation of the consequent parameters. The model building process comprises three stages. In stage one, structure learning is formulated as an objective weighting optimization problem. Apart from the mean square error (MSE) and the number of rules, three additional criteria are introduced in the fitness function for measuring the quality of the partitions. Optimization of these measures leads to models with representative rules, small overlapping and efficient data cover. To obtain models with good local interpretation, the consequent parameters are determined using a local MSE function while the overall model is evaluated on the basis of a global MSE function. The initial model is simplified at stage two using a rule base simplification routine. Similar fuzzy sets are merged and the “don’t care” premises are recognized. Finally, the simplified models are fine-tuned at stage three to improve the model performance. The suggested method is used to generate TSK models with crisp and polynomial consequents for two benchmark classification problems, the iris and the wine data. Simulation results reveal the effectiveness of our method. The resulting models exhibit simple structure, interpretability and superior recognition rates compared to other methods of the literature.     

Daily prediction of short-term trends of crude oil prices using neural networks exploiting multimarket dynamics

This paper documents a systematic investigation on the predictability of short-term trends of crude oil prices on a daily basis. In stark contrast with longer-term predictions of crude oil prices, short-term prediction with time horizons of 1–3 days posits an important problem that is quite different from what has been studied in the literature. The problem of such short-term predicability is tackled through two aspects. The first is to examine the existence of linear or nonlinear dynamic processes in crude oil prices. This sub-problem is addressed with statistical analysis involving the Brock-Dechert-Scheinkman test for nonlinearity. The second aspect is to test the capability of artificial neural networks (ANN) for modeling the implicit nonlinearity for prediction. Four experimental models are designed and tested with historical data: (1) using only the lagged returns of filtered crude oil prices as input to predict the returns of the next days; this is used as the benchmark, (2) using only the information set of filtered crude oil futures price as input, (3) combining the inputs from the benchmark and second models, and (4) combing the inputs from the benchmark model and the intermarket information. In order to filter out the noise in the original price data, the moving averages of prices are used for all the experiments. The results provided sufficient evidence to the predictability of crude oil prices using ANN with an out-of-sample hit rate of 80%, 70%, and 61% for each of the next three days’ trends.     

The FreeBSD project: a replication case study of open source development

Case studies can help to validate claims that open source software development produces higher quality software at lower cost than traditional commercial development. One problem inherent in case studies are external validity - we do not know whether or not results from one case study apply to another development project. We gain or lose confidence in case study results when similar case studies are conducted on other projects. This case study of the FreeBSD project, a long-lived open source project, provides further understanding of open source development. The paper details a method for mining repositories and querying project participants to retrieve key process information. The FreeBSD development process is fairly well-defined with proscribed methods for determining developer responsibilities, dealing with enhancements and defects, and managing releases. Compared to the Apache project, FreeBSD uses 1) a smaller set of core developers - developers who control the code base - that implement a smaller percentage of the system, 2) a larger set of top developers to implement 80 percent of the system, and 3) a more well-defined testing process. FreeBSD and Apache have a similar ratio of core developers to people involved in adapting and debugging the system and people who report problems. Both systems have similar defect densities and the developers are also users in both systems.     

多模态引导的局部特征选择小样本学习方法

深度学习模型取得了令人瞩目的成绩,但其训练依赖于大量的标注样本,在标注样本匮乏的场景下模型表现不尽人意.针对这一问题,近年来以研究如何从少量样本快速学习的小样本学习被提了出来,方法主要采用元学习方式对模型进行训练,取得了不错的学习效果.但现有方法:1)通常仅基于样本的视觉特征来识别新类别,信息源较为单一; 2)元学习的使用使得模型从大量相似的小样本任务中学习通用的、可迁移的知识,不可避免地导致模型特征空间趋于一般化,存在样本特征表达不充分、不准确的问题.为解决上述问题,将预训练技术和多模态学习技术引入小样本学习过程,提出基于多模态引导的局部特征选择小样本学习方法.所提方法首先在包含大量样本的已知类别上进行模型预训练,旨在提升模型的特征表达能力;而后在元学习阶段,方法利用元学习对模型进行进一步优化,旨在提升模型的迁移能力或对小样本环境的适应能力,所提方法同时基于样本的视觉特征和文本特征进行局部特征选择来提升样本特征的表达能力,以避免元学习过程中模型特征表达能力的大幅下降;最后所提方法利用选择后的样本特征进行小样本学习.在MiniImageNet、CIFAR-FS和FC-100这3个基准数...     

An improved model-based evolutionary algorithm for multi-objective optimization

The basic idea in the estimation of distribution algorithms is the replacement of heuristic operators with machine learning models such as regression models, clustering models, or classification models. So, recently, the model-based evolutionary algorithms (MBEAs) have been suggested in three groups: The estimation of distribution algorithms (EDAs), surrogate assisted evolutionary algorithms, and the inversed models to map from the objective space to the decision space. In this article, a new approach, based on an inversed model of Gaussian process and random forest framework, is proposed. The main idea is applying the process of random forest variable importance with a random grouping that determines some of the best assignment of decision variables to objective functions in order to form a Gaussian process in inverse models that maps to decision space the rich solutions which are discovered from objective space. Then these inverse models through sampling the objective space generate offspring. The proposed algorithm has been tested on the benchmark test suite for evolutionary algorithms (modified Deb K, Thiele L, Laumanns M, Zitzler E (DTLZ), and Walking Fish Group (WFG)) and indicates that the proposed method is a competitive and promising approach.     

The Train Benchmark: cross-technology performance evaluation of continuous model queries

In model-driven development of safety-critical systems (like automotive, avionics or railways), well-formedness of models is repeatedly validated in order to detect design flaws as early as possible. In many industrial tools, validation rules are still often implemented by a large amount of imperative model traversal code which makes those rule implementations complicated and hard to maintain. Additionally, as models are rapidly increasing in size and complexity, efficient execution of validation rules is challenging for the currently available tools. Checking well-formedness constraints can be captured by declarative queries over graph models, while model update operations can be specified as model transformations. This paper presents a benchmark for systematically assessing the scalability of validating and revalidating well-formedness constraints over large graph models. The benchmark defines well-formedness validation scenarios in the railway domain: a metamodel, an instance model generator and a set of well-formedness constraints captured by queries, fault injection and repair operations (imitating the work of systems engineers by model transformations). The benchmark focuses on the performance of query evaluation, i.e. its execution time and memory consumption, with a particular emphasis on reevaluation. We demonstrate that the benchmark can be adopted to various technologies and query engines, including modeling tools; relational, graph and semantic databases. The Train Benchmark is available as an open-source project with continuous builds from https://github.com/FTSRG/trainbenchmark.     

Parallel evolution strategy on grids for the protein threading problem

The protein threading problem is the problem of determining the three-dimensional structure of a given but arbitrary protein sequence from a set of known structures of other proteins. This problem is known to be NP-hard and current computational approaches to threading are unrealistic for long proteins and/or large template data sets. In this paper, we propose an evolution strategy for the solution of the protein threading problem. We also propose three parallel methods for fast threading. Our experiments produced encouraging preliminary results in term of threading energy as well as significant reduction in threading time.