分布事务管理器的性能优化

作为应用服务器重要组成部分的分布管理器主要用于提供分布环境下的事务处理能力,它的性能是制约应用服务器上分布事务吞吐能力的重要因素.讨论了在J2EE应用服务器框架下如何进行分布事务管理器的性能优化,同时给出了分布事务管理器ISTX的性能优化实现.实验数据表明,对ISTX的性能优化能在保证分布事务可靠性的前提下,较大幅度地提高了分布事务处理的性能.

短消息指标新定义及在事务信道限制中的应用

短消息指标可以度量隐蔽信道的短消息传输能力,是信道容量的必要补充.但指标现有定义中还存在着以下问题:消息长度参数在普通信息系统中不能定量分析;信道限制机制难以同时满足传输时间和保真度两个约束;没有包含消息的敏感度信息.针对这些问题,首先通过引入短消息传输价值的概念,给出了短消息指标的新定义.在该定义中,利用价值阈值统一表示系统对信道短消息传输能力的容忍程度,并且在所采用的价值函数中引入了消息的敏感度因素.其后,基于安全实时数据库应用场景给出了结合短消息指标和信道容量的事务隐蔽信道度量和限制机制.理论分析和模拟结果表明,基于短消息指标的新定义,系统可以对隐蔽信道威胁实施全面的度量和可调节的限制.

移动数据库系统乐观事务处理策略

在移动数据库系统中 ,计算平台的移动性、频繁的断接性以及长事务等特性使得传统的事务处理模式不再适用 .为了解决移动数据库中的事务处理问题 ,提出了一种新的移动事务处理模型——乐观两阶段提交移动事务模型 ( O2 PC-MT) .该模型采用乐观并发控制与两阶段提交协议相结合的方法 ,对移动事务的长事务特性提供了灵活与有效的支持 ;此外 ,该模型允许移动计算机分多次发送事务操作 ,且在事务执行的过程中可以任意移动 ,从而提供了对交互式事务及随意移动性的支持 .实验结果表明 ,与基于两段锁协议及其变形的其它移动事务处理模型相比 ,O2 PC-MT提高了系统的事务吞吐率并改善了系统的总体性能

可集成的制造执行系统

制造执行系统作为企业上层事务处理和低层设备控制系统之间的中间桥梁 ,在生产计划的优化调度、生产过程的改进等方面起着越来越重要的作用 ,并得到工业界广泛的认可和应用。本文总结了制造执行系统的发展历程和存在的问题 ,简要介绍了制造执行系统的功能、信息交互关系 ,以及信息技术对制造执行系统的冲击 ,进而引入工作流管理技术、智能Agent和面向对象技术 ,提出了可集成制造执行系统的体系结构、运行机制和开发方法 ,并对该系统的市场进行了分析与预测。

Synthesis of a group of diosgenyl saponins with combined use of glycosyl trichloroacetimidate and thioglycoside donors

With the combined use of glycosyl trichloroacetimidates and thioglycosides, a group of natural diosgenyl saponins (1-6) are efficiently synthesized, in either a stepwise or a 'one-pot' manner. The trichloroacetimidate is employed as an efficient temporary hydroxy protecting group in glycosylation with the glycosyl trichloroacetimidate. The intermolecular alkylthio-group transfer is demonstrated to be a common side-reaction during glycosylation with thioglycosides.

交易型领导的结构和作用机制研究

The Multifactor Leadership theory developed by Bass (1985) has become the new paradigm of leadership research. The empirical results of the effectiveness of transformational and transactional leadership in the literature, however, are not consentient. Researchers in China found the different structure of transformational leadership, but have not developed the transactional leadership. This study attempts to investigate three key questions in the unique Chinese socio-economic context: 1) what is the structure of transactional leadership in China? 2) What are the differences between western countries and China? And 3) what is the relationship between the transformational and transactional leadership mechanism? This study examines data collected from 3,500 participants, using Explored Factor Analysis (EFA), Confirmed Factor Analysis (CFA), Hierarchical Regression Analyses, partial correlations and other statistics methods. The major finings are listed as follows： Firstly, inductive methods was used to explore the structure of transactional leadership and the result show that transactional leadership is a four dimensions structure which includes contingent reward, contingent punishment , process control and anticipated investment. Reliability analysis, item analysis, EFA and CFA show the reliability and validity of the transactional leadership questionnaire we designed is good enough, the design of the item is effectively and properly. Contrast to other researches, anticipated investment emphasis on the leader’s recessive investment for subordinate, and this kind of transaction is quite special under the Chinese culture. While the content of the contingent reward with the contingent punishment is wider than the contingent reward in the western country, and the process control is wider than the management by exception and including goal setting and the management during the process. Secondly, hierarchical regression analyses showed that transformational and transactional leadership were significant positively related with in-role performance, extra-role performance, satisfaction and leadership effectiveness while negatively related to intention to leave. The effects of transactional and transformational leadership are different. Transactional leadership could significantly predict intention to leave controlling for transformational leadership, while transformational leadership could significantly predict in-role performance, extra-role performance, satisfaction and leadership effectiveness controlling for transactional leadership. Thirdly, the income level and the rank of subordinates are the moderators between the transformational, transactional leadership and leadership effectiveness. The leadership effectiveness of transactional leadership would decrease as the rank of subordinates increased, while the leadership effectiveness of transformational leadership would increase as the rank of subordinates increased. Transactional leadership is positively related to the effectiveness when the level of the subordinate income is low, but negatively related to the effectiveness when the level of the subordinate income is high. However the income level of the subordinate could not influence the leadership effectiveness of transformational leadership.

性别、效价及媒体对选择冲突问题解决的影响

Choice conflict is a critical issue in decision making. This study adopted stock transaction as the task to explore the relation between share price level and conflict intensity that the subjects experienced in decision making, and the effect of price level on individual and group choice conflict resolution. The strategy people used to solve choice conflict and gender difference were also examined. Modes of interaction (face to face and through audio media) in group decision making were also studied. The main results are as follows: 1. In individual decision-marking, there is a significant gender effect on decision time. Female subjects will spent more time on the task than male subjects. 2. When making a choice decision, the individual experienced stronger conflict for price shares than for low price shares. The stronger the conflict level they feel, the more difficult to make the choice decision. 3. Four strategies were used to finish the task. Male subjects used simple strategies while female subjects used more complex strategies. 4. In group decision-making, share price level had a significant effect on selection time. People used longer selection time for low price shares than for high price shares. No significant interaction was found between share price level and Modes of interaction. 5. Modes of interaction had significant effect on satisfaction coherence of the group. Under face-to-face condition, people within one group had greater satisfaction coherence. 6. Media had no significant effect on people's perception during the experiment. Satisfaction for cooperation, successful communication and good cooperation were correlated. Self satisfaction and satisfaction of the partners others were also correlated with the satisfaction of the whole task.