基于大值堆的自调整粗粒度并行遗传算法模型

一般粗粒度并行遗传算法（CGGA）的性能受诸多因素的影响表现不尽如人意。以降低通信代价为主要目标，受物种金字塔模型的启发，设计了一种双阈值限制下的自调整堆结构，并对其堆调整具体操作进行了改进，以期望改进后算法中种群间的通信代价大幅度降低，优化收敛速度，提高算法效率。通过对遗传算法的几个典型测试函数通信量的分析和实验表明，基于该模型的并行遗传算法在降低通信代价、提高收敛速度、优化最终解方面收效明显。     

旅行商问题(TSP)的伪并行遗传算法

旅行商问题(TSP)是典型的NP完全组合优化问题．本文基于遗传算法求解TSP问题时的独特性,提出一种采用无性繁殖的改进伪并行遗传算法,避免了交叉算子对良好基因模式的破坏;初始种群通过贪婪算法得到并进行预处理,提高算法的收敛速度;伪并行遗传算法中子群体之间的信息交换采用孤岛模型．这些改进措施对降低算法的复杂程度、提高算法的收敛速度和全局搜索能力有重要意义．仿真研究结果表明,该算法的寻优效率较高,有效地克服了标准遗传算法的早熟收敛问题．     

基于改进型遗传算法的多无人机任务分配

无人机集群的决策过程中存在个体任务分配问题，如在无人机集群执行侦察任务时存在的任务点分配问题，针对解决该问题时存在的计算时间长，搜索性能差等问题，提出一种针对算法机制和并发方式的改进方法。对无人机任务分配问题建立模型，将其抽象为多旅行商问题，通过遗传算法来解决，并将无人机实际飞行场景中的代价设置为动态和静态两种情况。在此基础上，通过改变计算适应度并行方式和添加缓存命中机制以减少算法冗余计算，提升算法适应度计算效率，采取优化的遗传算法淘汰机制使算法收敛速度提升。仿真结果验证，淘汰机制和并行缓存的优化使遗传算法具有更好的寻优能力和收敛速度，能够在较短的时间内规划出代价较小的航迹。     

基于多变异改进遗传算法的多agent系统通信

提出一种多变异改进遗传算法应用于多AGENT系统通信问题。考虑基因位的影响力及跟踪进化过程的遗传参数调整策略,采用多种变异操作,提高算法的收敛能力。仿真结果表明,改进遗传算法相对于SGA解决多AGENT问题在收敛速度上有明显提高,算法稳定性也有所改善。新算法的优化质量较高,具有更好的实际应用效果。     

一种种群自适应收敛的快速遗传算法

作为一种全局搜索算法,遗传算法的局部搜索能力较低,后期产生的无效进化与早熟收敛影响优化的速度和精度。已有的改进策略多以算法的时间复杂度为代价提高后期效率,严重限制了遗传算法在工业控制系统中的应用。针对这种情况,提出了一种新型种群自适应收敛的快速遗传算法,即通过提高种群的遗传质量,在严格控制算法复杂度的前提下提高优化性能。仿真结果证明,在不增加时间复杂度的前提下,新算法显著地提升了收敛精度和收敛速度。     

基于代价优化的云工作流调度改进PSO算法

云计算可以通过即付即用的方式向用户工作流提供资源。为了解决资源服务代价异构环境下的云工作流任务调度代价问题,提出一种基于改进粒子群算法的云工作流任务调度算法WSA-IPSO。通过综合考虑任务的执行代价和依赖任务间发生数据传输时的通信代价,算法将总代价优化问题形式化为有向无环图DAG中的任务调度模型,并提出基于改进粒子群算法的优化模型对其进行求解。通过改进传统粒子群算法的粒子速度更新策略和惯性权重更新策略,算法可以以更快的收敛速度得到代价最小化的调度方案。通过仿真实验,与MCT算法及标准粒子群算法进行性能比较。实验结果表明,WSA-IPSO算法在降低总代价、任务分布的负载均衡以及算法收敛性方面比较同类算法均表现出更好的性能。     

智能算法在全终端网络可靠性优化设计中的应用

在计算机通信网络建设过程中,通信主干网的高可靠性一直是广大设计者和用户所关心的重大问题,分析了通信主干网的全终端可靠性优化设计问题,针对一种典型可靠性优化模型,分别设计了用于求解该模型的改进遗传算法、模拟退火算法、改进蚁群算法以及改进粒子群算法,并对该模型进行了优化求解,计算机仿真实例表明:粒子群算法在算法的优化质量和收敛速度上都优于其它三种算法,能够取得满意的优化结果,具有一定的实用价值.     

微粒群优化算法在基站分布规划中的应用

微粒群优化算法（PSO）是一种具有全局随机优化的智能算法。该算法编程简单,鲁棒性强,并行运算能力强,能以较快的速度收敛至全局最优解。本文在PSO的思想基础上提出了一种改进搜索方向,降低“早熟”概率的方法,即结合遗传算法,引入了爬坡算子。本文并将该改进算法应用于基站分布规划的研究中,实验仿真结果表明,利用该改进算法能提高基站覆盖率,降低经济成本。     

一种多策略并行遗传算法及其收敛性分析

为解决传统遗传算法收敛速度慢、群体多样性不足的缺陷,提出了一种多策略并行的遗传算法;算法采用多策略并行处理的方式,产生不同策略模式下的个体,增加群体的多样性,再经过自适应迁移策略,提高算法的收敛速度;利用Markov链模型分析多策略并行遗传算法的收敛性;采用旅行商函数进行算法测试,结果表明改进算法的收敛性较传统遗传算法有较大的提高,具有较强的工程应用性能。     

具有量子行为粒子群优化算法的并行化研究

在研究了具有量子行为粒子群算法的基础上,受遗传算法并行化的启发,对具有量子行为的粒子群算法提出并实现了新的并行化策略。针对通信时间过长的问题,提出了改进方法。最后通过benchmark测试函数,将并行化量子粒子优化算法和二进制遗传算法、十进制遗传算法、粒子群优化算法的并行化方法进行了仿真比较,并对结果进行了分析。     

一种居住区位和通勤交通方式的组合选择模型

为了研究通勤交通方式对就业者选择居住区位的影响,基于交叉分层的Logit模型建立了居住区位和通勤交通方式的组合选择模型。以不同通勤距离为半径作居住区位选择环域集,以通勤交通方式和居住区位的两两组合为模型选择项集合,以就业者社会经济属性、居住成本、通勤成本为特征变量建立概率选择模型。根据调查数据,采用阶段计算法对特征变量和模型参数进行标定和检验。测试结果表明:通勤时间、通勤费用和居住成本与选择项关系密切;收入较高且拥有小汽车的就业者对于居住区位选择的灵活性较大。     

A communication-reduced and computation-balanced framework for fast graph computation

The bulk synchronous parallel (BSP) model is very user friendly for coding and debugging parallel graph algorithms. However, existing BSP-based distributed graph-processing frameworks, such as Pregel, GPS and Giraph, routinely suffer from high communication costs. These high communication costs mainly stem from the fine-grained message-passing communication model. In order to address this problem, we propose a new computation model with low communication costs, called LCC-BSP. We use this model to design and implement a high-performance distributed graph-processing framework called LCC-Graph. This framework eliminates high communication costs in existing distributed graph-processing frameworks. Moreover, LCC-Graph also balances the computation workloads among all compute nodes by optimizing graph partitioning, significantly reducing the computation time for each superstep. Evaluation of LCC-Graph on a 32-node cluster, driven by real-world graph datasets, shows that it significantly outperforms existing distributed graph-processing frameworks in terms of runtime, particularly when the system is supported by a high-bandwidth network. For example, LCC-Graph achieves an order of magnitude performance improvement over GPS and GraphLab.     

P-3PC: a point-to-point communication model for automatic andoptimal decomposition of regular domain problems

One of the most fundamental problems automatic parallelization tools are confronted with is to find an optimal domain decomposition for a given application. For regular domain problems (such as simple matrix manipulations), this task may seem trivial. However, communication costs in message-passing programs often depend significantly on the memory layout of data blocks to be transmitted. As a consequence, straightforward domain decompositions may be non-optimal. In this paper, we introduce a new point-to-point communication model, called P-3PC (Parameterized model based on the Three Paths of Communication), that is specifically designed to overcome this problem. In comparison with related models (e.g. LogGP), P-3PC is similar in complexity, but more accurate in many situations. Although the model is aimed at MPI's standard point-to-point operations, it is applicable to similar message-passing definitions as well. The effectiveness of the model is tested in a framework for automatic parallelization of low-level image processing applications. Experiments are performed on two Beowulf-type systems, each having a different interconnection network and a different MPI implementation. The results show that, where other models frequently fail, P-3PC correctly predicts the communication costs related to any type of domain decomposition     

六边形网格细分法实现二阈值推算定位

推算定位技术是大规模分布交互仿真系统缓解网络传输延时及减少网络通讯量的传统方法。以往的推算定位技术在设定推算阈值时 ,总因为顾及近距离实体的精确度需求而设置较小的阈值 ,这样无可避免地增加了远距离实体的数据接收频率 ,增加了网络通讯开销。为此本文提出了多阈值推算定位技术 ,针对不同距离的接收实体对应不同的误差容许阈值进行推算 ,针对 DIS系统给出了一种基于六边形网络细分的二阈值推算定位技术的实现方法。该方法实现简单 ,可以有效减少系统的通讯量     

Cluster partition-based communication of multiagents: The model and analyses

In previous works, the agent communication always takes place randomly through the whole agents distributed area, and all agents are treated equally in the communication process. However, the real communication is always imbalanced: communication in some places of the system may be very busy and the communication resources are insufficient, whereas in other places the communication may be very idle and the communication resources are redundant. To make up the shortage of related works, this paper presents a cluster partition-based communication model for multiagents. In the presented model, agents are divided into some clusters according to their communication frequencies; in each cluster, one agent is set to be hub-agent which carries the information of other agents within the cluster. With the presented model, the communication costs within the same cluster is lower than the one among different clusters because agents have closer distances within the same cluster; moreover, many real communication processes may take place within clusters since the clusters are partitioned according to the communication frequencies among agents. Therefore, the communication time and costs can be reduced by adopting our model. Finally, the advantage of our model is proved by a series of experiments.     

Psychological costs of support seeking and choice of communication channel

While seeking support brings benefits, it also entails some costs to the seeker. We propose that seeking support involves two types of psychological costs: intra- and inter-personal costs. Intra-personal cost is defined as the psychological threat arising from the perception that one fails to achieve one's own aspiration, while inter-personal cost is defined as the psychological threat arising from the perception that one fails to meet others' expectation. These costs result from individuals focusing on different aspects of the self and will deter individuals from seeking support. We conducted two studies. In the first study, we adapted, developed and validated scales to measure the two types of psychological costs. In the second study, we examined the impact of both types of psychological costs on individuals' choice of communication channel when seeking support. We found that only inter-personal cost was significantly related to seekers' choice of communication channel. Specifically, the higher the inter-personal cost perceived, the greater the likelihood that individuals would prefer email over face-to-face communication. We also found that women were more likely to seek support through email than face-to-face communication compared to men. Implications of the results for research and practice are discussed.     

An Operational Semantics for Shared Messaging Communication

Shared Messaging Communication (SMC) has been introduced in [Satya Kiran M.N.V., Jayram M.N., Pradeep Rao, and S.K. Nandy. A complexity effective communication model for behavioral modeling of signal processing applications. In Proceedings of 40th Design Automation Conference, 2003] as a model of communication which reduces communication costs (both in terms of communication latency and memory usage) by allowing tasks to communicate data through special shared memory regions. Sending a reference to an otherwise inaccessible memory regions rather than the data itself, the model combines the advantages of message passing and shared memories. Experimental results have shown that SMC in case of large data payloads clearly outperforms the classical message passing.In this paper we give a formal operational semantics to SMC exhibiting unambiguously the effect of executing an SMC command on local and shared memories. Based on this semantics we show that any program using message passing can be proved to be weakly bisimilar to one based on SMC and that with respect to communication costs the latter is amortised cheaper, [A. Kiehn and S. Arun-Kumar. Amortised bisimulations. In Proceedings of FORTE 2005, number 3731 in Lecture Notes in Computer Science, pages 320–334. Springer-Verlag, 2005].     

Application-driven,energy-efficient communication in wireless sensor networks

Several sensor network applications based on data diffusion and data management can determine the communication transfer rate between two sensors beforehand. In this framework, we consider the problem of energy efficient communication among nodes of a wireless sensor network and propose an application-driven approach that minimizes radio activity intervals and prolongs network lifetime. On the basis of possible communication delays we estimate packet arrival intervals at any intermediate hop of a fixed-rate data path. We study a generic strategy of radio activity minimization wherein each node maintains the radio switched on just in the expected packet arrival intervals and guarantees low communication latency. We define a probabilistic model that allows the evaluation of the packet loss probability that results from the reduced radio activity. The model can be used to optimally choose the radio activity intervals that achieve a certain probability of successful packet delivery for a specific radio activity strategy. Relying on the probabilistic model we also define a cost model that estimates the energy consumption of the proposed strategies, under specific settings. We propose three specific strategies and numerically evaluate the associated costs. We finally validate our work with a simulation made with TOSSIM (the Berkeley motes’ simulator). The simulation results confirm the validity of the approach and the accuracy of the analytic models.     

基于Agent的分布式数据库查询优化研究

目前分布式数据库查询优化器基于代价模型,总代价=CPU代价+I/O代价+通信代价,在远程通信网或数据传输率较低的系统中,通信代价应作为首要的优化目标来考虑;提出了一种基于Multi-Agent技术的查询模型(QMAS);同时为了保证系统组内、组间Agent之间的高效通信,采用了树型拓扑结构(TTMAS)的通信模型,系统内各Agent使用通信原语高效通信、协作,且满足Agent间的通信路由最优,从而保证了QMAS的查询效率.     

LoGPG: Modeling network contention in message-passing programs

In many real applications, for example, those with frequent and irregular communication patterns or those using large messages, network contention and contention for message processing resources can be a significant part of the total execution time. This paper presents a new cost model, called LoGPC, that extends the LogP and LogGP models to account for the impact of network contention and network interface DMA behavior on the performance of message passing programs. We validate LoGPC by analyzing three applications implemented with Active Messages on the MIT Alewife multiprocessor. Our analysis shows that network contention accounts for up to 50 percent of the total execution time. In addition, we show that the impact of communication locality on the communication costs is at most a factor of two on Alewife. Finally, we use the model to identify trade-offs between synchronous and asynchronous message passing styles