论数字图书馆系统中WEB服务器性能优化的方法

由于在数字图书馆系统中流通着的大多是数字化的索引、文摘、全文、图像或音频视频等多媒体信息,对Web服务器性能有着较高的要求.结合实际工程的经验,从硬件实现手段(缓存服务器、均衡负载设备、Web双机镜像等)和软件实现手段(多层软件结构设计、应用部署等)等两个大方面论述如何提高Web服务器性能,以便使用户能够更快捷、高效、安全地使用数字图书馆应用系统.     

ProxyServer中Cache的管理和使用

使用Ｃａｃｈｅ可以减少Ｂｒｏｗｅｒ的等待时间和降低代理服务器和Ｉｎｔｅｒｎｅｔ的通信量,本文主要介绍了Ｐｒｏｘｙ对Ｃａｃｈｅ管理和使用。     

透过专利看微处理器的技术发展（九）——多核处理器中CaChe专利技术分析

本文在对近几年多核处理器中高速缓存（Cache）相关专利进行深入全面的调查基础上,以Intel公司的专利为样本,分析了微处理器进入多核时代后Cache技术研究重点。对多核中Cache关键技术的发展趋势进行了展望,希望能为产品研发提供线索和思路。     

改进的基于目录的Cache一致性协议

介绍几种典型目录一致性协议并分析它们的优缺点。在综合全映射目录和有限目录优点的基础上,通过在存储器层上增加一个存储器高速缓存（Cache）层的方式,提出并讨论一种改进后的Cache一致性协议。该协议相对有限目录存储开销增加不多的情况下,提高了系统性能和可扩展性。     

高档微处理器结构的发展

本文综述了32位微处理器结构的发展情况及提高处理器性能所采取的策略手段,对比采用精简指令系统 RISC 设计和复杂指令系统CISC设计的两类不同微处理器的性能特点,同时论述了32位专用微处理器芯片的结构。     

Analysis and reduction of data spikes in thin client computing

While various optimization techniques have been used in existing thin client systems to reduce network traffic, the screen updates triggered by many user operations will still result in long interactive latencies in many contemporary network environments. Long interactive latencies have an unfavorable effect on users’ perception of graphical interfaces and visual contents. The long latencies arise when data spikes need to be transferred over a network while the available bandwidth is limited. These data spikes are composed of a large amount of screen update data produced in a very short time. In this paper, we propose a model to analyze the packet-level redundancy in screen update streams caused by repainting of graphical objects. Using this model we analyzed the data spikes in screen update streams. Based on the analysis result we designed a hybrid cache-compression scheme. This scheme caches the screen updates in data spikes on both server and client sides, and uses the cached data as history to better compress the recurrent screen updates in possible data spikes. We empirically studied the effectiveness of our cache scheme on some screen updates generated by one of the most bandwidth-efficient thin client system, Microsoft Terminal Service. The experiment results showed that this cache scheme with a cache of 2M bytes can reduce 26.7%–42.2% data spike count and 9.9%–21.2% network traffic for the tested data, and can reduce 25.8%–38.5% noticeable long latencies for different types of applications. This scheme costs only a little additional computation time and the cache size can be negotiated between the client and server.     

Runtime home mapping for effective memory resource usage

In tiled Chip Multiprocessors (CMPs) last-level cache (LLC) banks are usually shared but distributed among the tiles. A static mapping of cache blocks to the LLC banks leads to poor efficiency since a block may be mapped away from the tiles actually accessing it. Dynamic policies either rely on the static mapping of blocks to a set of banks (D-NUCA) or rely on the OS to dynamically load pages to statically mapped addresses (first-touch).     

企业应用中的缓存统计模式

首先讨论企业应用中数据访问和性能的重要性,缓存和缓存统计的必要性。然后提出了一种缓存统计的实现,指出其不足。并提出了基于面向对象的设计原则实现的缓存统计模式,与简单的缓存统计方法进行了比较。     

一种改进的监视Cache协议算法

本文探讨ＳＭＰ体系结构的多处理环境下,几种常见Ｃａｃｈｅ一致性问题的解决方案,讨论和对比了各自的特色,在此基础上进行一步分析了基中的不足,并提出我们的方案－－改进的监视Ｃａｃｈｅ协议算法。