减少Hadoop集群中网络队头阻塞的调度算法

大数据分析系统的用户希望任务的执行时间尽可能短.然而,在任务执行期间,网络与计算时刻都可能成为阻碍任务执行的资源瓶颈.通过对大数据分析系统的观察与分析,得出如下结论:1)根据当前资源瓶颈的不同,数据并行框架应当在多种工作模式之间切换;2)子任务的调度应当充分考虑将来可能到达的新任务,而不能仅考虑当前已经提交的任务.基于...     

Check-in services and passenger behaviour: Self service technologies in airport systems

The need for airports to optimise space and the need for airlines to bring down their operating costs favours the use of self-service technologies in services provided to passengers and at check-in, specifically. Checking in online and at kiosks in the airport is gaining ground on the use of the airline check-in desk. The objective of this paper is to analyse the socio-demographic factors or flight characteristics that influence a passenger’s choice of check-in from the various options available. For this a multinomial logit is used and applied to an extensive sample of almost 20,000 passengers, of whom 43% were foreigners, at five Spanish airports. The factors that determine the choice of check-in mode include the passenger’s age and level of education, the reason for making the journey, waiting time and the type of airline. The universal use of the new technologies in airport management, and the broad cosmopolitan sample mean that the conclusions can be easily extrapolated to other airport systems.     

Proportionate progress: A notion of fairness in resource allocation

Given a set ofn tasks andm resources, where each taskx has a rational weightx.w=x.e/x.p,0<1, aperiodic schedule is one that allocates a resource to a taskx for exactlyx.e time units in each interval [x.p·k, x.p·(k+1)) for allkN. We define a notion of proportionate progress, called P-fairness, and use it to design an efficient algorithm which solves the periodic scheduling problem.This research was supported by NSF Research Initiation Award CCR-9111591, and the Texas Advanced Research Program under Grant No. 91-003658-480.     

History-dependent scheduling: Models and algorithms for scheduling with general precedence and sequence dependence

In this paper, we extend job scheduling models to include aspects of history-dependent scheduling, where setup times for a job are affected by the aggregate activities of all predecessors of that job. Traditional approaches to machine scheduling typically address objectives and constraints that govern the relative sequence of jobs being executed using available resources. This paper optimises the operations of multiple unrelated resources to address sequential and history-dependent job scheduling constraints along with time window restrictions. We denote this consolidated problem as the general precedence scheduling problem (GPSP). We present several applications of the GPSP and show that many problems in the literature can be represented as special cases of history-dependent scheduling. We design new ways to model this class of problems and then proceed to formulate it as an integer program. We develop specialized algorithms to solve such problems. An extensive computational analysis over a diverse family of problem data instances demonstrates the efficacy of the novel approaches and algorithms introduced in this paper.     

Controller Area Network (CAN) schedulability analysis: Refuted,revisited and revised

Controller Area Network (CAN) is used extensively in automotive applications, with in excess of 400 million CAN enabled microcontrollers manufactured each year. In 1994 schedulability analysis was developed for CAN, showing how worst-case response times of CAN messages could be calculated and hence guarantees provided that message response times would not exceed their deadlines. This seminal research has been cited in over 200 subsequent papers and transferred to industry in the form of commercial CAN schedulability analysis tools. These tools have been used by a large number of major automotive manufacturers in the design of in-vehicle networks for a wide range of cars, millions of which have been manufactured during the last decade. This paper shows that the original schedulability analysis given for CAN messages is flawed. It may provide guarantees for messages that will in fact miss their deadlines in the worst-case. This paper provides revised analysis resolving the problems with the original approach. Further, it highlights that the priority assignment policy, previously claimed to be optimal for CAN, is not in fact optimal and cites a method of obtaining an optimal priority ordering that is applicable to CAN. The paper discusses the possible impact on commercial CAN systems designed and developed using flawed schedulability analysis and makes recommendations for the revision of CAN schedulability analysis tools. Robert I. Davis received a DPhil in Computer Science from the University of York in 1995. Since then he has founded three start-up companies, all of which have succeeded in transferring real-time systems research into commercial product. At Northern Real-Time Technologies Ltd. (1995–1997) he was responsible for development of the Volcano CAN software library. At LiveDevices Ltd. (1997–2001) he was responsible for development of the Real-Time Architect suite of products, including an OSEK RTOS and schedulability analysis tools. In 2002, Robert returned to the University of York, and in 2004 he was involved in setting up a spin out company, Rapita Systems Ltd., aimed at transferring worst-case execution time analysis technology into industry. Robert is a member of the Real-Time Systems Research Group at the University of York, and a director of Rapita Systems Ltd. His research interests include scheduling algorithms and schedulability analysis for real-time systems. Alan Burns is head of the Real-Time Systems Research Group at the University of York. His research interests cover a number of aspects of real-time systems including the assessment of languages for use in the real-time domain, distributed operating systems, the formal specification of scheduling algorithms and implementation strategies, and the design of dependable user interfaces to real-time applications. He has authored/co-authored over 370 papers and 10 books, with a large proportion of them concentrating on real-time systems and the Ada programming language. Professor Burns has been actively involved in the creation of the Ravenscar Profile, a subset of Ada”s tasking model, designed to enable the analysis of real-time programs and their timing properties. Reinder J. Bril received a B.Sc. and an M.Sc. (both with honours) from the University of Twente, and a Ph.D. from the Technische Universiteit Eindhoven, the Netherlands. He started his professional career in January 1984 at the Delft University of Technology. From May 1985 until August 2004, he was with Philips, and worked in both Philips Research as well as Philips’ Business Units. He worked on various topics, including fault tolerance, formal specifications, software architecture analysis, and dynamic resource management, and in different application domains, e.g. high-volume electronics consumer products and (low volume) professional systems. In September 2004, he made a transfer back to the academic world, to the System Architecture and Networking (SAN) group of the Mathematics and Computer Science department of the Technische Universiteit Eindhoven. His main research interests are currently in the area of reservation-based resource management for networked embedded systems with real-time constraints. Johan J. Lukkien has been head of the System Architecture and Networking Research group at Eindhoven University of Technology since 2002. He received an M.Sc. and a Ph.D. from Groningen University in the Netherlands. In 1991, he joined Eindhoven University, after two years leave at the California Institute of Technology. His research interests include the design and performance analysis of parallel and distributed systems. Until 2000 he was involved in large-scale simulations in physics and chemistry. Since 2000, his research focus has shifted to the application domain of networked resource-constrained embedded systems. Contributions of the SAN group are in the area of component-based middleware for resource-constrained devices, distributed co-ordination, Quality of Service in networked systems and schedulability analysis in real-time systems.     

Taking advantage of a diverse set of efficient production schedules: A two-step approach for scheduling with side concerns

In many practical scheduling problems, the concerns of the decision-maker may not be all known in advance and therefore may not be included in the initial problem definition as an objective function and/or as constraints. In such a case, the usual techniques of multi-objective optimization become inapplicable. To cope with this problem and to facilitate handling the concerns of the decision-maker, which can be implicit or qualitative, a dedicated methodological framework is needed. In this paper we propose a new two-step framework. First, we aim at obtaining a set of schedules that can be considered efficient with respect to a performance measure and at the same time different enough from one another to enable flexibility in the final choice. We formalize this new problem and suggest to address it with a multimodal optimization approach. Niching considerations are discussed for common scheduling problems. Through the flexibility induced with this approach, the additional considerations can be taken into account in a second step, which allows decision-makers to select an appropriate schedule among a set of sound schedules (in contrast to common optimization approaches, where usually a single solution is obtained and it is final). The proposed two-step approach can be used to handle a wide range of underlying scheduling problems. To show its potential and benefits we illustrate the framework on a set of hybrid flow shop instances that have been previously studied in the literature. We develop a multimodal genetic algorithm that employs an adapted version of the restricted tournament selection for niching purposes in the first step. The second step takes into account additional concerns of the decision-maker related to the ability of the schedules to absorb the negative effects due to random machine breakdowns. Our computational experiments indicate that the proposed framework is capable of generating numerous high-performance (mostly optimal) schedules. Additionally, our computational results demonstrate that the proposed framework provides the decision-maker a high flexibility in dealing with subsequent side concerns, since there are drastic differences in the capabilities of the efficient solutions found in Step 1 to absorb the negative impacts of machine breakdowns.     

Multimodal express shipment service design: Models and algorithms

The research objective is to develop models and algorithms to provide quality solutions for large-scale service design problems. Service design problems arising at railroads, airlines, trucking firms, intermodal partnerships, etc. require the determination of the cost minimizing or profit maximizing set of services and their schedules, given limited resources and service requirements. The model is applied to a large express shipment transportation problem involving over 1.3 billion decision variables and 200,000 constraints. We develop a new model and solution approach — branch-and-price-and-cut. Computational results show that near optimal solution is achieved within a reasonable run time using novel problem reduction methods involving node consolidation, link consolidation, derived schedules and a branch-and-price-and-cut solution procedure.     

Post disaster adaptation management in airport: A coordination of runway and hangar resources for relief cargo transports

Air traffic congestions for processing relief cargos under post-disaster relief scenarios are common, due to high transport demands within a short time. To enhance the resilience of relief operations at airport, an optimization problem of relief air cargo transportations involving aircraft sequencing and loading/unloading within a designated hangar is studied in this paper. The objective is minimizing the tardiness in fulfilling inbound and outbound relief cargos. A mixed-integer linear programming model is formulated, which incorporates aircraft sequencing and hangar parking planning. To resolve the practical problem efficiently, we propose a two-stage optimization approach, which reduces complexity in solving the original model by coordinating the decisions of aircraft landing and take-off schedule and cargo hangar parking arrangement through iterations. The efficiency of the proposed method is examined through the computational results. High-quality solutions can be obtained by the two-stage optimization method within a reasonable time for practical implementation, which enhances the responsiveness and resource utilization of airport operations management under disaster relief situations.     

Cyclic routing algorithms in graphs: Performance analysis and applications to robot scheduling

In this paper we revisit and extend the algorithm for the cyclic project scheduling problem which was originally proposed by Romanovskii (1967). While the algorithm has been derived for fixed numerical data, we show how it can be extended to handle the problems with interval data. We also propose a new algorithm for the cyclic scheduling problem with interval data that extends the parametric method developed by Megiddo (1979) and runs in strongly polynomial time.     

MANET移动模型分析与仿真实验

根据随机性、时间相关、空间相关、地理受限等特性对于MANET移动模型进行了分类研究,结合3种实际应用,包括战场环境、事件场景、参观展览,比较移动模型与之相应的特点和适用范围,研究了移动模型混合使用的可行性。利用ANSIM工具对其中常用的几种移动性模型进行仿真试验,比较了RW、RD、GM模型对节点平均邻居数、平均跳数、节点间连通性的影响,并进行了校园区内障碍物模型下节点移动的场景模拟。     

General Flowshop Models: Job Dependent Capacities, Job Overlapping and Deterioration

Several extensions of the two-machine flowshop model that arise in industrial applications are considered. These modifications of the classical case concern buffer capacities that vary with the size of the part, and processing times that are variable and increase in situations of overlapping and long waiting periods between the different operations.     

基于时间序列分析的航站楼安检旅客流量预测

对单位时间内通过安检的旅客流量进行预测是机场航站楼实时调控的重要依据,由此提出一种实时安检旅客流量预测方法,采用Wolf方法分析出安检旅客流量时间序列具有混沌特性;采用适用于混沌时间序列预测的遗传算法优化BP神经网络预测方法(GABP)预测安检旅客流量;分别设定时间尺度为2 min、5 min和10 min等,分析不同时间尺度对安检旅客流量预测精度的影响。基于北京首都国际机场T3航站楼实际安检旅客流量数据的实验结果表明,采用GABP神经网络对以2 min为时间尺度的安检旅客流量预测能取得较好的预测精准度。     

Agility and production flow layouts: An analytical decision analysis

The term ‘agile manufacturing’ has referred to operational aspects of a manufacturing company concerning their ability to produce customized products at mass production prices and with short lead times. A core issue faced within agile manufacturing is the need for appropriate and supporting production and operations systems. Many design dimensions of agility and agile manufacturing exist. To help attain this goal for integrating the many design dimensions, operations infrastructure and capacity must be carefully planned to manage production flow, and thus production layout planning takes on an increasingly important role. Given the importance of these dimensions in response to agility, this paper seeks to make a contribution by providing insights into a decision aid for evaluating production flow layouts that support and enhance the agile manufacture of products. Layout design has a significant impact on the performance of a manufacturing or service industry system and has been an active research area for many decades. Strategic evaluation of production layouts requires consideration of both qualitative and quantitative factors (managerial, organizational, and technical). This paper makes use of the Analytical Network Process (ANP) which captures interdependencies among different criteria, sub-criteria and dimensions, an evident characteristic of production flow layouts in complex agile manufacturing environments. An application case study exemplifying the practical usefulness of this type of model describes how management, after implementation of the model, made a mid-course correction related to the production layout initially selected.     

分布式网络安全扫描任务调度模型及算法

通过分析影响分布式网络安全扫描任务调度的主要因素，从提高扫描效率的角度提出了一种新的扫描任务调度模型和调度算法。该模型已实现，算法的有效性在实践中得到了检验。     

圩区泵闸群自动化调度控制策略研究

为实现圩区泵闸群控制的自动化、信息化,提出完整的自动化系统设计方案。为降低泵闸群常规调度的能耗,提出一套基于正交试验法的调度策略优化运行方案,通过9次试验获得不同开机方案和运行规则下的总能耗,再进行排序和去劣,选出最优调度方案。以上海市新浜镇圩区为例,阐述优化的完整计算过程,优化后能够比常规调度方案节约能源6%以上。提升圩区的自动化水平和排涝效率,有效节约能源,同时方案的研究方法具有一定的实用和参考价值。     

铁路客运站客流组织的仿真与优化

通过研究铁路客运站行人的路径选择行为,以排队长度、楼梯高度、携带行李量、紧急程度和年龄等因素为路径选择的参数,对车站内行人在楼梯与自动扶梯之间选择的影响机理进行分析,采用二值Logistic回归方法建立了行人对楼梯与自动扶梯的选择行为模型。在对客运站的客流组织仿真方法和评价方法研究的基础上,针对客运站客流组织存在的问题,运用AnyLogic仿真软件建立客流的微观仿真模型。实例分析验证了该方法的有效性和可行性,为设计、优化车站设施布局和车站客流组织,提高运营安全、可靠性和服务水平提供了参考依据。     

Hierarchical facility location problem: Models,classifications, techniques,and applications

The primary objective in a typical hierarchical facility location problem is to determine the location of facilities in a multi-level network in a way to serve the customers at the lowest level of hierarchy both efficiently (cost minimization objective) and effectively (service availability maximization objective). This paper presents a comprehensive review of over 40 years of hierarchical facility location modeling efforts. Published models are classified based on multiple characteristics including the type of flow pattern, service availability, spatial configuration, objective function, coverage, network levels, time element, parameters, facilities, capacity, and real world application. A second classification is also presented on the basis of solution methods adopted to solve various hierarchical facility location problems. The paper finally identifies the gaps in the current literature and suggests directions for future modeling efforts.     

Online-review analysis based large-scale group decision-making for determining passenger demands and evaluating passenger satisfaction: Case study of high-speed rail system in China

High-speed rail (HSR) has become an essential mode of public transportation in China and is likely to remain so for the foreseeable future. To promote the development of the HSR industry, a high level of passenger satisfaction must be ensured, which means that passenger satisfaction must be assured. Focusing on HSR in-cabin factors that affect the travel experience of HSR passengers, this study aims to determine passenger demands (PDs) and to evaluate passenger satisfaction by using a combination of online review analysis and large-scale group decision-making (LSGDM). By using web crawler technology, online reviews related to HSR were harvested from a microblogging platform to extract PD data and information. The six PDs that reflect the most frequent concerns of passengers were identified by analyzing the online reviews. The level of satisfaction of passengers with respect to these PDs was analyzed based on the online responses from 100 HSR passengers and by adopting the interval-valued two-tuple linguistic representation model. The final degrees of satisfaction and rankings of the PDs were then determined by using the LSGDM approach with the k-means clustering method and a consensus-reaching process. This research thus constructs an index system of HSR passenger satisfaction evaluation based on online-review analysis and evaluates the process by using LSGDM approaches. The conclusions provide insights into the improvements desired by HSR passengers for in-cabin services and to improve passenger satisfaction.