供热外网采用直埋式保温管的讨论

从工程投资,施工及保温材料的技术性能的分析入手,认为采用硬脂聚氨酯泡沫塑料保温的直埋管道比地沟管更具有优越性。     

Performance analysis of long-lived transaction processing systemswith rollbacks and aborts

Increasing the parallelism in transaction processing and maintaining data consistency appear to be two conflicting goals in designing distributed database systems (DDBSs). This problem is especially difficult if the DDBS is serving long-lived transactions (LLTs). A special case of LLTs, called sagas, has been introduced that addresses this problem. A DDBS with sagas provides high parallelism to transactions by allowing sagas to release their locks as early as possible. However, it is also subject to an overhead, due to the efforts needed to restore data consistency in the case of failure. We conduct a series of simulation studies to compare the performance of LLT systems with and without saga implementation in a faulty environment. The studies show that saga systems outperform their nonsaga counterparts under most of conditions, including heavy failure cases. We thus propose an analytical queuing model to investigate the performance behavior of saga systems. The development of this analytical model assists us to quantitatively study the performance penalty of a saga implementation due to the failure recovery overhead. Furthermore, the analytical solution can be used by system administrators to fine-tune the performance of a saga system. This analytical model captures the primary aspects of a saga system, namely data locking, resource contention and failure recovery. Due to the complicated nature of the analytical modeling, we solve the model approximately for various performance metrics using decomposition methods, and validate the accuracy of the analytical results via simulations     

Neuronal units linked to microvascular modules in cerebral cortex: response elements for imaging the brain

How neuronal activity changes cerebral blood flow is of biological and practical importance. The rodent whisker-barrel system has special merits as a model for studies of changes in local cerebral blood flow (LCBF). Stimulus-evoked changes in neural firing and 'intrinsic signals' recorded through a cranial window were used to define regions of interest for repeated flow measurements. Whisker-activated changes in flow were measured with intravascular markers at the pia. LCBF changes were always prompt and localized over the appropriate barrel. Stimulus-related changes in parenchymal flow monitored continuously with H2 electrodes recorded short latency flow changes initiated in middle cortical layers. Activation that increased flow to particular barrels often led to reduced flow to adjacent cortex. Dye was injected into single penetrating arterioles from the pia of the fixed brain and injected into arterioles in slices of cortex where barrels were evident without stains. Arteriolar and venular domains at the surface were not directly related to underlying barrels. Capillary tufts in layer IV were mainly coincident with barrels. The matching between a capillary plexus (a vascular module) and a barrel (a functional neuronal unit) is a spatial organization of neurons and blood vessels that optimizes local interactions between the two. The paths of communication probably include: neurons to neurons, neurons to glia, neurons to vessels, glia to vessels, vessels to vessels and vessels to brain. Matching a functional grouping of neurons with a vascular module is an elegant means of reducing the risk of embarrassment for energy-expensive neuronal activity (ion pumping) while minimizing energy spent for delivery of the energy (cardiac output). For imaging studies this organization sets biological limits to spatial, temporal and magnitude resolution. Reduced flow to nearby inactive cortex enhances local differences.     

专用设备驱动程序交互式生成器的设计与实现

主要作为电子高科技的重点领域，专用实时设备无论类别和型号，抑或数量和质量，均与日俱增，令人目不暇接。它们的高效管理与灵活加接，是任何计算机专用实时系统不可或缺的基本功能，其驱动模块随之成为任何专用实时系统无法回避的设计难点。本文介绍我们自行研制的专用实时设备驱动程序交互式生成器ＲＴＤＧ的设计与实现。     

Adrenergic regulation of aqueous outflow

Epinephrine increased outflow facility and cyclic AMP in the in vitro perfused human anterior segment with a maximal facility increase of 44% occurring at approximately 2 x 10(-5) M. Cyclic AMP measured in the perfusate from anterior segments increased by 12-14 fold after administration of 10(-5) M epinephrine. Both the facility increase and cyclic AMP rise were blocked by the beta-2 selective antagonist, ICI118,551. While there was a correlation between the facility increase and elevation in cyclic AMP levels, the rise in cyclic AMP preceded the facility increase by about 1 hour, suggesting that the ultimate effect of epinephrine involved a rather slow event such as synthesis and release of prostaglandins or protein synthesis. Subsequent perfusion studies showed that very large concentrations of indomethacin were necessary to block the outflow facility effect of epinephrine, suggesting that prostaglandin synthesis did not underlie the facility effect in this system. However, 5 x 10(-5) M cyclohexamide blocked the effect on outflow facility of both epinephrine and forskolin, but did not block the rise in cyclic AMP. These studies suggest that protein synthesis may play a role in the epinephrine-induced facility increase at some point beyond the second messenger level.     

Finding a maximum matching in a permutation graph

We present anO(n log logn) time algorithm for finding a maximum matching in a permutation graph withn vertices, assuming that the input graph is represented by a permutation.     

MOCVD生长的CdZnTe／ZnTe多量子附的激子光学特性

用常压ＭＯＣＶＤ方法在ＧａＡｓ（１００）衬底上生长了ＣｄＺｎＴｅ／ＺｎＴｅ多量子阱。在室温下，观测到了ＣｄＺｎＴｅ／ＺｎＴｅ多量子阱的三个谱带发光。根据ＣｄＺｎＴｅ／ＺｎＴｅ多量子阱的吸收光谱和不同激发光强下的发光光谱，分别归结ＣｄＺｎＴｅ／ＺｎＴｅ多量子阱中观测到的三个发光谱带于覆盖层发光、ｎ＝１的重空穴激子发光及杂质发光。     

基于超高频雷达的湖面风浪参数提取

基于Barrick的微绕散射理论,在超高频雷达USCR工作频率条件下,将反演均方高浪高公式应用于超高频雷达USCR提取湖面风浪信息,在国内首次实现从湖面回波多普勒谱中获取了浪高信息;并利用Heron理论,由浪高信息反演出风速风向信息;通过模拟相同风浪状态下的湖面回波,并进行对比证明本方法是行之有效的。     

一种自适应的仿生图像增强方法:LDRF算法

为了研究基于人类视觉系统特性(亮度自适应特性和视网膜神经元感受野非经典侧抑制特性)的仿生图像处理方法,增强仿生图像增强算法的自适应性,提出了一种新的自适应仿生图像增强算法——LDRF算法.LDRF算法首先建立参数对数模型对图像全局亮度进行自适应调整,然后采用三高斯动态滤波进行局部细节增强,引入Wal-lis算子建立增益因子模型,使局部细节增强具有自适应性,最后通过线性变换恢复图像彩色信息.在大量图像上进行对比实验和分析.实验结果证明,LDRF算法能够避免过增强现象,并且针对不同大小、不同内容的图像能够自适应地进行图像增强,取得了较好的效果,提高了仿生图像增强算法的实用性.