基于Jini的集群作业管理系统的实现

集群作业管理系统对于集群的使用很重要,本文研究如何基于Jini技术构建集群作业管理系统-Jcluster。文章首先介绍Jini技术和该技术的特点,然后阐述了基于Jini技术的Jcluster的设计思想和系统结构,并在自强2000集群系统上实现。经过测试,证实了该系统的具有即插即用的特性和良好的健壮性、同时Jcluster实现了对资源的多级资源池管理。     

On-line scheduling on an unbounded parallel batch machine to minimize makespan of two families of jobs

We study the on-line scheduling on an unbounded parallel batch machine to minimize makespan of two families of jobs. In this model, jobs arrive over time and jobs from different families cannot be scheduled in a common batch. We provide a best possible on-line algorithm for the problem with competitive ratio . Research supported by NSFC (10671183), NFSC-RGC (70731160633) and SRFDP (20070459002).     

Optimisation of PV-wind-diesel-battery stand-alone systems to minimise cost and maximise human development index and job creation

In this paper we show a multi-objective evolutionary algorithm (MOEA) for the optimisation of stand-alone (off-grid) hybrid systems (photovoltaic-wind-diesel-battery) to minimise total net present cost (NPC) and maximise human development index (HDI) and job creation (JC). Optimisation of this kind of system is usually performed considering only the minimisation of cost (NPC or the levelised cost of energy), as well as the emissions and the unmet load in some cases. In this paper, for the first time, we consider the maximisation of HDI and JC as part of optimisation. HDI depends on the consumption of electricity, so the extra energy that can supply the hybrid system can improve the HDI index. JC is different for each technology, obtaining different values for each combination of components in the system. The three objectives are often opposed, so a Pareto-optimisation MOEA is a good option to obtain a set of possible solutions in which no solution is better than another one for all three objectives (optimal Pareto set). We provide an example in the optimisation of a hybrid system to supply electricity to a small community in the Sahrawi refugee camps of Tindouf.     

An integrated Multi-Regional Input-Output (MRIO) Analysis of miscanthus biomass production in France: Socio-economic and climate change consequences

Several European policies have been designed over the last decades to address the challenge of climate change and several measures have been put in place to accelerate the development and deployment of cost-effective low carbon technologies. The domestic nature of the resource and its great potential availability in Europe make biomass conversion technologies relevant mitigation options to be considered. In this context, the project “Logistics for Energy Crops Biomass (LogistEC)” aims to develop new or improve technologies of biomass logistics chain. In this project, the sustainability of different types of biomass is analysed in terms of environmental, economic and social impacts, based on the supply chain of two existing plants. The objective of this paper is to present the main results obtained in the socio-economic analysis of the French case and its climate change consequences. The Input-Output Analysis (IOA) has been seen as the most appropriate method to estimate these impacts using a Multiregional Input-Output Table from the World Input-Output Database project. Socio-economic effects have been estimated in terms of additional economic activity, added value and job creation. By extending the IOA with environmental accounts, greenhouse gas (GHG) emissions have also been estimated. Additionally, the most stimulated sectors have been identified. Results highlight the importance of biomass at a national level.     

Cell phones during nonwork time: A source of job demands and resources

Technology (e.g., cell phones) is increasingly blurring the lines between the work and nonwork domains. Evidence suggests technology users experience both negative and positive outcomes associated with work-related technology use during nonwork hours. We extended the job demands-resources model (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001) to technology use by conceptualizing work-related cell phone (WRCP) use as a job demand and cell phone attachment –valuing and being physically attached to a cell phone– as a resource. We expected high cell phone attachment will buffer against the negative effects of WRCP use on emotional exhaustion, work engagement, and work-family conflict. Participants from various occupations (N = 313) responded to two online surveys administered one week apart. Cell phone use and attachment were assessed at Time 1; criteria were assessed at Time 2. High cell phone attachment buffered against the negative effects of WRCP use on emotional exhaustion and work-family conflict, and it enhanced the beneficial effects of WRCP use on work engagement. Being more engaged and attached to cell phones may help employees deal with WRCP use during nonwork time more effectively. Practical implications include providing training for more effective cell phone use during nonwork time.     

Workers' intrinsic work motivation when job demands are high: The role of need for autonomy and perceived opportunity for blended working

Work overload or work pressure may undermine workers' intrinsic motivation. In the present research, we tested the conditions under which this may (not) occur, including the perceived opportunity to blend on-site and off-site working through the effective use of computers and modern information and communication technology. Our sample consisted of 657 workers (51% female) representing a variety of industries. As hypothesized, it is not high job demands per se, but high demands in combination with a high need for autonomy and a lack of perceived opportunities for blended working that undermines intrinsic work motivation. When workers high in need for autonomy perceived opportunities for blended working, their intrinsic work motivation was not negatively affected by increasing job demands. This main finding suggests that, particularly for workers high in need for autonomy, the perceived opportunity for blended working is an effective, contemporary resource to cope with the increasing job demands typically observed in today's workplace. Theoretically, these findings contribute to the refinement and extension of influential demands-resource models and Person-Job Fit theory. Practically, our findings may show managers how to effectively keep workers intrinsically motivated and productive in their jobs when job demands are high.     

A study on local search neighborhoods for the job shop scheduling problem with total weighted tardiness objective

In this paper we consider the job shop scheduling problem with total weighted tardiness objective (JSPTWT). This objective reflects the goal to achieve a high service level which is of increasing importance in many branches of industry. The paper concentrates on a class of baseline heuristics for this problem, known as neighborhood search techniques. An approach based on disjunctive graphs is developed to capture the general structure of neighborhoods for the JSPTWT. Existing as well as newly designed neighborhoods are formulated and analyzed. The performance and search ability of the operators (as well as combinations thereof) are compared in a computational study. Although no dominant operator is identified, a transpose-based perturbation on multiple machines turns out as a promising choice if applied as the only operator. Combining operators improves the schedule quality only slightly. But, the implementation of operators within a meta-heuristic enables to produce a higher schedule quality. A structural classification of neighborhood operators and some new analytical results are presented as well.     

求解较大规模JSSP的自适应混合遗传算法

针对一类以最小化加权拖期时间为调度目标的Job Shop调度问题（JSSP），提出一种自适应混合遗传算法。首先，在遗传算法迭代求解过程中，为降低调度问题的求解规模，基于所定义的调度特征量——资源冲突可能性，将所有操作动态划分为资源冲突可能性较高的操作和资源冲突可能性较低的操作，分别直接和间接参与染色体编码。然后，基于上述划分，遗传算法中的染色体由直接参与编码的操作序列构成的基因串、表示启发式规则的基因串（用于确定间接参与染色体编码的操作的加工优先顺序）和标志串3段基因串组成。另外，构造了一个模糊逻辑控制器用于自适应调节第一段基因串的长度，以提高算法性能。数值仿真结果表明，在求解一类较大规模的JSSP时所提算法是有效的。     

一种快速网格任务调度策略

网格任务调度目标有很多,如用户要求任务轮转时间短、花费代价小,而资源提供者希望资源利用率高等,这些目标相互冲突,因此网格任务调度不仅是一个NP难问题,而且是一个多目标优化问题.本文根据网格环境下任务的时间相关性特点,对传统蚁群算法进行了改进,提出了一种快速网格任务调度算法.该算法不仅解决了网格调度中多目标优化问题,而且依据任务调度历史信息生成蚁群算法的初始信息素分布,提高了蚁群算法的求解速度.     

网格环境下的作业运行支持系统分析*

网格环境下的作业运行支持系统支持用户在网格资源上远程提交作业任务,执行科学计算应用程序,并管理运行着的作业任务.作业运行支持系统解决了计算执行环境的准备、状态监视汇报、运行时操纵和I/O支持等方面的关键问题.现有的几种主要的网格中间件系统均提供了作业执行和管理工具,很好地解决了几个主要问题,但并不能完全满足用户的需要,还需进一步改进与完善.