iSCSI高可用存储解决方案的设计和实现

介绍了心跳传输机制和iSCSI协议的原理和实现。以成熟的开源软件Hearbeat和 UNH iSCSI软件为基础,在Linux操作系统平台上设计和实现了一套在广域网下基于iSCSI技术的高可用存储解决方案,从而为中小企业关键业务提供经济、高可用的存储服务。     

iSCSI的性能分析和优化

本文从iSCSI协议读写流程的角度。对iSCSI协议各个部分的服务时间进行量化分析，讨论了它们对iSCSI协议性能的影响。提出了相应的优化策略。同时，本文讨论和分析了iSCSI Initiator模块的队列长度对iSCSI协议性能的影响。提出了改进方法。     

基于iSCSI存储集群的设计与实现

为了节约成本,提高系统的可用性,通过研究iSCSI与存储集群的特点,提出将iSCSI应用到存储集群中的方案.分析了心跳检测、故障迁移、数据镜像等技术,在设计心跳检测时,通过增加失效点阀值提高检测的正确性,故障迁移利用ARP欺骗原理来完成,数据镜像则使用同步模式来保障数据的一致性.最后实现了一款高可靠、高可用基于iSCSI的存储集群系统,并对存储系统性能进行了测试.     

双机高可用系统设计与性能分析

针对融合iSCSI,NAS,SAN的海量网络存储系统的特点,设计了一种双机高可用元数据服务器系统,该系统不仅减少了元数据服务器瓶颈,而且可充分保证存储网络系统元数据的高可用性.通过建立连续时间马尔可夫链性能分析模型,分析结果显示双机高可用系统的可用度远优于单机单路径系统.     

Linux下基于iSCSI存储系统的实现与性能评测

本文分析了新的网络存储协议--iSCSI协议，并给出了Linux 下 iSCSI协议实现的一种软件模型，进而搭建于iSCSI的存储系统，对其进行了性能测试，并指出了优化iSCSI系统性能的途径。     

Linux下iSCSI存储系统中逻辑卷管理的设计与实现

iSCSI存储是一种新型的存储技术，将逻辑卷管理模块应用于iSCSI存储系统，可以提高系统的性能和可靠性。文章研究了RAID技术以及iSCSI存储系统的原理和工作流程，提出了Linux系统下iSCSI存储系统中逻辑卷管理的设计方法，并提供了一套参考实现。     

iSCSI,CIFS,NFS协议的性能评测

在Windows系统下对比测试了iSCSI协议与CIFS协议在文件访问上的性能；在Linux系统下对比测试了iSCSI协议与NFS协议在文件访问上的性能．得出了iSCSI协议在文件访问上优于CIFS和NFS协议，且适应于海量存储的结论，并分析了其原因．     

基于IPSec的分级安全iSCSI技术研究

分析了iSCSI面临的安全问题，根据不同的安全需求，设计了一个基干IPSec的分级安全iSCSI方案，并通过试验对不同级别安全方案的性能进行评估。试验结果表明，在高配置硬件环境下，安全iSCSI具有较好的性能；在低配置硬件环境上，安全iSCSI会占用大量的系统资源，并且吞吐量也大幅度降低。     

iSCSI协议的仿真实现

分析了iSCSI协议原理，提出了利用iSCSI协议规定的特性和参数在OPNET8.0上实现iSCSI协议的思路和方法，建立了iSCSI仿真模型，分析了SCSI和iSCSI层的实现方法，定义了在实现过程中所处的状态，阐述了这两层在OPNET中的实现过程。     

一种改善iSCSI存储系统性能的缓冲策略

iSCSI是一种在IP网络上传输SCSI协议的协议。然而，在IP网络上封装SCSI协议需要大量额外开销来进行SCSI命令的传输和网络中的握手动作。通过对iSCSI协议及美国UNH(university of new hampshire)大学的参考代码实现进行分析，提出了一种改善iSCSI存储系统的性能的缓冲策略。通过对缓冲区进行有效的管理可以缩短系统的响应时间，并提高系统的并行程度，从而可以达到改善iSCSI存储性能的目的。     

More experience might not bring more safety: Negative moderating effect of pilots’ flight experience on their safety performance

ObjectiveThis study aimed to improve our understanding of the relationship between pilots' total flight hours (TFH), levels of hazardous attitude (HA), and flight safety performance.BackgroundThe negative effect of airline pilots' flight experience on safety performance has received attention but has not been extensively investigated.MethodUsing a sample of 45 Chinese commercial airline pilots, we not only tested the mediation model in which the relationship between TFH and exceedance rate was mediated by HA, but also tested the moderation model in which the relationship between HA and exceedance rate was moderated by TFH.ResultsTFH positively predict flight exceedance rates; more importantly, it moderated the effect of HA on flight exceedance rates. Specifically, TFH exacerbated the negative effect of HA on flight exceedance rates. Meanwhile, the mediating effect of HA was not significant.ConclusionResults implies the nonlinear relationship between TFH, HA, and flight safety performance. The moderation model indicates that the experience-related risks may appear before 5000 h for an individual commercial airline pilot.ApplicationResearchers, safety managers, and policymakers in the aviation industry that conduct pilots' psychological competency research in the Professionalism Lifecycle Management (PLM) system should be aware of the potentially interactive effects of hazardous attitude and experience on commercial airline pilots’ flight safety performance.     

Hybrid adaptive fuzzy identification and control of nonlinearsystems

We present a combined direct and indirect adaptive control scheme for adjusting an adaptive fuzzy controller, and adaptive fuzzy identification model parameters. First, using adaptive fuzzy building blocks, with a common set of parameters, we design and study an adaptive controller and an adaptive identification model that have been proposed for a general class of uncertain structure nonlinear dynamic systems. We then propose a hybrid adaptive (HA) law for adjusting the parameters. The HA law utilizes two types of errors in the adaptive system, the tracking error and the modeling error. Performance analysis using a Lyapunov synthesis approach proves the superiority of the HA law over the direct adaptive (DA) method in terms of faster and improved tracking and parameter convergence. Furthermore, this is achieved at negligible increased implementation cost or computational complexity. We prove a theorem that shows the properties of this hybrid adaptive fuzzy control system, i.e., bounds for the integral of the squared errors, and the conditions under which these errors converge asymptotically to zero are obtained. Finally, we apply the hybrid adaptive fuzzy controller to control a chaotic system, and the inverted pendulum system     

一种增强服务器高可用性的设计方法

首先简要介绍了HA(High Availability)的内容和工作原理,然后详细分析了HA的心跳数据格式和心跳流程,给出了一种增强服务器高可用性的设计方法,并在Unix操作系统Solaris上采用C 完成了HA模块开发.     

Transforming and tracing reused requirements models to home automation models

ContextModel-Driven Software Development (MDSD) has emerged as a very promising approach to cope with the inherent complexity of modern software-based systems. Furthermore, it is well known that the Requirements Engineering (RE) stage is critical for a project’s success. Despite the importance of RE, MDSD approaches commonly leave textual requirements specifications to one side.ObjectiveOur aim is to integrate textual requirements specifications into the MDSD approach by using the MDSD techniques themselves, including metamodelling and model transformations. The proposal is based on the assumption that a reuse-based Model-Driven Requirements Engineering (MDRE) approach will improve the requirements engineering stage, the quality of the development models generated from requirements models, and will enable the traces from requirements to other development concepts (such as analysis or design) to be maintained.MethodThe approach revolves around the Requirements Engineering Metamodel, denominated as REMM, which supports the definition of the boilerplate based textual requirements specification languages needed for the definition of model transformation from application requirements models to platform-specific application models and code.ResultsThe approach has been evaluated through its application to Home Automation (HA) systems. The HA Requirement Specification Language denominated as HAREL is used to define application requirements models which will be automatically transformed and traced to the application model conforming to the HA Domain Specific Language.ConclusionsAn anonymous online survey has been conducted to evaluate the degree of acceptance by both HA application developers and MDSD practitioners. The main conclusion is that 66.7% of the HA experts polled strongly agree that the automatic transformation of the requirements models to HA models improves the quality of the HA models. Moreover, 58.3% of the HA participants strongly agree with the usefulness of the traceability matrix which links requirements to HA functional units in order to discover which devices are related to a specific requirement. We can conclude that the experts we have consulted agree with the proposal we are presenting here, since the average mark given is 4 out of 5.     

用于协作信息处理的基组织结构模型

文中提出了大型互联网络上支持复杂协作策略的基组织结构模型（ＢａｓｉｃＯｒａｇｎｉｚａｔｉｏｎＳｔｒｕｃｔｕｒｅ,简称,ＢＯＳ）,在基组结构模型中,一个协作信息系统可分解为多个基本的子系统,第一基本的子系统用一个基组织来模型化。一个基组织内部一般定义五类主体;基组织协调控制主体,基组织维护管理主体,问题求解主体,过程性知识集成主体和公共设施主体,基组织内部的协作行为通过各类主体分工合作完成,其组织具     

高可用性系统中实时“监听”的解决

高可用性系统中的每一台服务器必须实时地“监听”对等机的状态,发现问题时能够立即自动接替其工作,文章描述了基于UNIX平台的高可用性系统中实时控制和实现的方法。     

Extended state observer-based motion synchronisation control for hybrid actuation system of large civil aircraft

Hybrid actuation system with dissimilar redundant actuators, which is composed of a hydraulic actuator (HA) and an electro-hydrostatic actuator (EHA), has been applied on modern civil aircraft to improve the reliability. However, the force fighting problem arises due to different dynamic performances between HA and EHA. This paper proposes an extended state observer (ESO)-based motion synchronisation control method. To cope with the problem of unavailability of the state signals, the well-designed ESO is utilised to observe the HA and EHA state variables which are unmeasured. In particular, the extended state of ESO can estimate the lumped effect of the unknown external disturbances acting on the control surface, the nonlinear dynamics, uncertainties, and the coupling term between HA and EHA. Based on the observed states of ESO, motion synchronisation controllers are presented to make HA and EHA to simultaneously track the desired motion trajectories, which are generated by a trajectory generator. Additionally, the unknown disturbances and the coupling terms can be compensated by using the extended state of the proposed ESO. Finally, comparative simulation results indicate that the proposed ESO-based motion synchronisation controller can achieve great force fighting reduction between HA and EHA.     

Fault tolerance for home agents in mobile IP

In the conventional IP protocol, whenever a mobile device moves to a different network, it must change its IP address to communicate with other nodes in the Internet. Mobile IP protocol keeps mobile nodes online without changing their IP addresses while changing the attachment points. The packets destined to mobile nodes (MN) are relayed by their home agents (HA). However, once the only one HA fails, all MNs managed by the HA will not receive packets normally. In the paper, we propose a novel protocol with multiple mobile agents (MA) where only double mobility bindings are maintained in the whole system. When an HA is failed, its backup HA can take over it in a short time without fetching the bindings from other places. Besides, we also consider the load balancing between these HAs during HA takeover and recovery. Through the simulation and analyses, we found that our method has less registration overheads, better MN-scalability, less sensitivity on MN mobility, more fault-tolerance robustness, and less takeover time than others.     

Towards highly available and scalable high performance clusters

In recent years, we have witnessed a growing interest in high performance computing (HPC) using a cluster of workstations. This growth made it affordable to individuals to have exclusive access to their own supercomputers. However, one of the challenges in a clustered environment is to keep system failure to the minimum and to achieve the highest possible level of system availability. High-Availability (HA) computing attempts to avoid the problems of unexpected failures through active redundancy and preemptive measures. Since the price of hardware components are significantly dropping, we propose to combine both HPC and HA concepts and layout the design of a HA-HPC cluster, considering all possible measures. In particular, we explore the hardware and the management layers of the HA-HPC cluster design, as well as a more focused study on the parallel-applications layer (i.e. FT-MPI implementations). Our findings show that combining HPC and HA architectures is feasible, in order to achieve HA cluster that is used for High Performance Computing.     

开放式环境中的高可用性系统

高可靠性正在为高档次的开放系统环境中不可缺少的特性。