公开(公告)号：US20090106381A1

公开(公告)日：2009-04-23

申请号：US12341746

申请日：2008-12-22

申请人： Stephane Kasriel ,  Xavier Casanova ,  Walter Mann

发明人： Stephane Kasriel ,  Xavier Casanova ,  Walter Mann

IPC分类号： G06F15/16

CPC分类号： G06F17/30902 ,  H04L67/02 ,  H04L67/28 ,  H04L67/2847

摘要： A network object predictive pre-download device operable on a network logically between a web client and a web server on a network is described. The pre-download device includes a prediction engine that intercepts a request from the web client to the web server for a first network object and determines a second network object most probable to be requested in the future from the web server. The pre-download device also includes a cache primer that intercepts a response to the request from the web server and sends a first message to the web client informing the web client of the second network object to be cached by the web client. The cache primer also sends the second network object to the web client such that the second network object is predicatively pre-downloaded to the web client whereby second network object latency is reduced when an actual user request occurs.

摘要翻译： 描述了在网络上在网络客户机和网络服务器之间逻辑地在网络上可操作的网络对象预测预先下载设备。 预先下载设备包括预测引擎，该预测引擎拦截来自网络客户端的对于第一网络对象的Web服务器的请求，并且确定将来可能从web服务器请求的第二网络对象。 预下载设备还包括缓存引用，其拦截来自web服务器的对请求的响应，并向Web客户端发送第一消息，通知web客户端要由Web客户端缓存的第二网络对象。 缓存引用还将第二网络对象发送到web客户端，使得第二网络对象被预先预先下载到web客户端，从而当发生实际用户请求时第二网络对象延迟被减少。

公开(公告)号：US20070198687A1

公开(公告)日：2007-08-23

申请号：US11679158

申请日：2007-02-26

申请人： Stephane Kasriel ,  Walter Mann

发明人： Stephane Kasriel ,  Walter Mann

IPC分类号： G06F15/173

CPC分类号： H04L67/2814 ,  H04L67/2842 ,  H04L67/288

摘要： A system and method for delivering content on a network using differential caching is described. A request for information is received from a request-generating element coupled to a browser associated with a client through the network. A static portion and a dynamic portion of a document to be included in a response as the requested information are identified based on previously cached information by utilizing a software element that is not part of the client. The static portion is cached in a memory that is logically local to the client that requested the information. The static portion is served to the request-generating element from the cached memory. The dynamic portion is served to the request-generating element from the network. The static portion and the dynamic portion are integrated using the request-generating element into the document included in the response to the browser as the requested information.

摘要翻译： 描述了使用差分缓存在网络上传送内容的系统和方法。 从通过网络与与客户端相关联的浏览器耦合的请求生成元件接收到对信息的请求。 通过利用不是客户端的一部分的软件元素，基于先前缓存的信息来识别作为所请求的信息被包括在响应中的文档的静态部分和动态部分。 静态部分缓存在逻辑上位于请求信息的客户端的内存中。 静态部分从缓存的存储器提供给请求生成元素。 动态部分从网络提供给请求生成元素。 将静态部分和动态部分使用请求生成元素集成到包括在对浏览器的响应中的文档中作为所请求的信息。

公开(公告)号：US07188214B1

公开(公告)日：2007-03-06

申请号：US09923809

申请日：2001-08-07

申请人： Stephane Kasriel ,  Walter Mann

发明人： Stephane Kasriel ,  Walter Mann

IPC分类号： G06F12/06

CPC分类号： H04L69/04 ,  G06F17/30902 ,  H04L67/02 ,  H04L67/28 ,  H04L67/2842 ,  H04L67/2885

摘要： A technique for increased efficiency of content delivery over a network is described. Instances of web pages are divided into (1) templates including those elements of a web page that are relatively unchanging and (2) delta information including those elements that are ephemeral or customized. Each template is compressed and cached at an originating server. Transmission of the delta information is decoupled from transmission of the template. When a user requests a page, the compressed template is sent (either from an originating server or a mirror thereof). The delta information is compressed and sent separately. Since the template is only compressed once and is cached locally, it requires less bandwidth and allocation of other computing resources to transmit than transmission of a compressed web page.

摘要翻译： 描述了通过网络提高内容传送效率的技术。 网页的实例分为（1）个模板，包括网页的相对不变的元素，（2）增量信息，包括那些短暂或定制的元素。 每个模板在始发服务器上进行压缩和缓存。 增量信息的传输与模板的传输分离。 当用户请求页面时，压缩的模板被发送（来自始发服务器或其镜像）。 增量信息被压缩并单独发送。 由于模板只被压缩一次并且在本地进行缓存，所以与传送压缩网页相比，需要更少的带宽和其他计算资源的分配。

公开(公告)号：US07185063B1

公开(公告)日：2007-02-27

申请号：US09888374

申请日：2001-06-22

申请人： Stephane Kasriel ,  Walter Mann

发明人： Stephane Kasriel ,  Walter Mann

IPC分类号： G06F15/16

CPC分类号： H04L67/2814 ,  H04L67/2842 ,  H04L67/288

摘要： The invention provides a technique for serving relatively non-static content in a CDN (content delivery network). Content is pushed out from an originating server to a set of mirroring servers using differential caching. Each requestable object is separated into template and delta information. The originating server determines templates for differential caching, and causes those templates to be distributed using the CDN. Each mirroring server in the CDN is able to access a copy of a recent template for each request. Hosting of the template information is decoupled from hosting of the delta information. Delta information can include (a) changes to the template information since last distributed, or (b) personalized information specific to the client or to the request. Delta information can be served separately from the originating server, or can be served separately from mirroring servers using a CDN (possibly the same CDN as used to serve template information).

摘要翻译： 本发明提供了一种用于在CDN（内容传送网络）中服务相对非静态内容的技术。 内容使用差分缓存从始发服务器推送到一组镜像服务器。 每个可请求对象分为模板和增量信息。 始发服务器确定差分缓存的模板，并使用CDN分发这些模板。 CDN中的每个镜像服务器都可以访问每个请求的最近的模板副本。 托管模板信息与主机的差异信息脱钩。 增量信息可以包括（a）自上次分发以来对模板信息的更改，或（b）客户端特定的个性化信息或请求。 增量信息可以与始发服务器分开服务，或者可以使用CDN（可能与用于提供模板信息的CDN相同的CDN）与镜像服务器分开提供服务。

公开(公告)号：US20060075068A1

公开(公告)日：2006-04-06

申请号：US11262225

申请日：2005-10-28

申请人： Stephane Kasriel ,  Xavier Casanova ,  Walter Mann

发明人： Stephane Kasriel ,  Xavier Casanova ,  Walter Mann

IPC分类号： G06F15/16

CPC分类号： G06F17/30902 ,  H04L67/02 ,  H04L67/28 ,  H04L67/2847

摘要： A method for fetching network objects from a web server is described. The method includes intercepting at a device a request from a web client to a web server for a first network object. The same device intercepts a response to the request from the web server. The device determines at least one second network object most probable to be requested in the future from the web server. Subsequently, a message is sent from the device to the web client the first message informing the web client of the at least one second network object to be cached by the web client.

摘要翻译： 描述了一种从Web服务器获取网络对象的方法。 该方法包括在设备处拦截来自web客户端的请求到针对第一网络对象的web服务器。 相同的设备拦截来自Web服务器的对请求的响应。 该设备确定将来从web服务器最可能请求的至少一个第二网络对象。 随后，从设备向web客户端发送消息，通知Web客户端要被Web客户端缓存的至少一个第二网络对象的第一消息。

公开(公告)号：US07269784B1

公开(公告)日：2007-09-11

申请号：US09827268

申请日：2001-04-04

申请人： Stéphane Kasriel ,  Walter Mann

发明人： Stéphane Kasriel ,  Walter Mann

IPC分类号： G06N3/00

CPC分类号： G06F17/30902

摘要： The invention provides a method and system for sending relatively identical web pages, when requested by subsequent users, with substantial reduction of bandwidth. The server determines a “template web page” corresponding to the actual information on the web page, and having a set of insertion points, at which changed data can be inserted by the client. The server sends a JavaScript program corresponding to the template web page, which makes reference to the template web page and the changed data, along with sending the actual changed data itself. A first user requesting the web page receives the entire web page, while a second user requesting the web page (or the first user re-requesting the web page at a later time) receives the template information plus only the changed data. The server re-determines the template web page from time to time, such as when a ratio of changed data to template web page data exceeds a selected threshold. The server identifies the particular template web page to the client using a unique identifier (an “E-tag”) for the particular data sent in response to the request. Since the E-tag refers to the template, not the underlying web page, when the standard client makes its conditional request for the web page “if not changed”, the server responds that the web page is “not changed” even if it really is, but embeds the changed data in a cookie it sends to the client with the server response to the client request.

摘要翻译： 本发明提供了一种方法和系统，用于在后续用户请求时发送相对相对的网页，带宽的实质性减少。 服务器确定与网页上的实际信息相对应的“模板网页”，并且具有一组插入点，在该插入点处可以由客户端插入更改的数据。 服务器发送与模板网页相对应的JavaScript程序，引用模板网页和更改的数据，以及发送实际更改的数据本身。 请求网页的第一用户接收整个网页，而请求网页（或第一用户稍后重新请求网页）的第二用户接收模板信息加上仅改变的数据。 服务器不时地重新确定模板网页，例如当更改的数据与模板网页数据的比例超过所选择的阈值时。 服务器使用针对请求发送的特定数据的唯一标识符（“E-tag”）向客户端标识特定模板网页。 由于电子标签是指模板，而不是底层网页，当标准客户端对网页进行条件请求（如果没有更改）时，服务器就会回应网页“没有改变”，即使真的 是，但将更改后的数据嵌入其发送到客户端的cookie，服务器响应客户端请求。

公开(公告)号：US06963874B2

公开(公告)日：2005-11-08

申请号：US10128594

申请日：2002-04-22

申请人： Stephane Kasriel ,  Sara Swanson ,  Walter Mann

发明人： Stephane Kasriel ,  Sara Swanson ,  Walter Mann

IPC分类号： G06Q30/02 ,  H04L29/08 ,  G06F17/30

CPC分类号： G06Q30/02 ,  G06F17/3089 ,  H04L67/24 ,  H04L69/329 ,  Y10S707/99945

摘要： A data collection system collects statistics related to each visit to each web-page within a web-site. In addition to collecting path information related to traversals to and from each web-page, the system collects information regarding the particular session, such as the visitor's connection speed, the time required to download the web-page to the visitor's system, the duration of time that the visitor spent at the page, whether the user clicked the ‘back’ button to leave the site, and so on. The data reporting system presents statistics based on the collected data in an easy-to-use and easy-to-understand form that is customizable by the user. By collecting and reporting performance factors related to each web-page, or sets of web-pages within a web-site, marketers and webmasters can address any perceived deficiencies and/or can optimize the performance of select pages to achieve specific marketing objectives. To optimize performance and minimize storage requirements, the data collection system uses a two-stage storage architecture. Data is collected in one or more registers in a sequential fashion, and transferred to a database periodically.

摘要翻译： 数据收集系统收集与网站内每个网页的每次访问相关的统计数据。 除了收集与每个网页的遍历相关的路径信息之外，系统收集有关特定会话的信息，例如访问者的连接速度，将网页下载到访问者系统所需的时间， 访问者在页面上花费的时间，用户是否点击“返回”按钮离开网站，等等。 数据报告系统基于收集的数据，以易于使用且易于理解的形式呈现用户可定制的统计信息。 通过收集和报告与每个网页相关的性能因素或网站中的网页集合，营销人员和网站管理员可以解决任何感知的缺陷和/或可以优化选择页面的性能以实现特定的营销目标。 为了优化性能和最小化存储要求，数据收集系统使用两阶段存储架构。 数据以顺序方式收集在一个或多个寄存器中，并且周期性地传送到数据库。

公开(公告)号：US4560708A

公开(公告)日：1985-12-24

申请号：US671444

申请日：1984-11-14

申请人： Peter Horn ,  Anton Hesse ,  Walter Mann ,  Wolfram Frank ,  Matthias Marx ,  Walter Heckmann

发明人： Peter Horn ,  Anton Hesse ,  Walter Mann ,  Wolfram Frank ,  Matthias Marx ,  Walter Heckmann

IPC分类号： C08G18/46 ,  C08G18/68 ,  C08G63/91 ,  C08G18/14 ,  C08G18/42

CPC分类号： C08G18/4669 ,  C08G18/68 ,  C08G63/918 ,  C08G2105/02 ,  Y10S521/902

摘要： Crystallite suspensions containing from 3 to 70 weight percent of a crystalline, ethylenically unsaturated polyester prepared through condensation polymerization, and having a molecular weight factor per double bond of 154.4 to 408, as the dispersed phase, and from 30 to 97 weight percent of a polyhydroxyl compound having a functionality of from 2 to about 8, a hydroxyl number of from 20 to 800, and a melting point of less than 30.degree. C. as the coherent phase are suitable for preparing non-cellular and cellular polyurethane- or polyurethane group-containing polyisocyanurate foams.

摘要翻译： 含有3〜70重量％的通过缩聚制备的结晶性烯键式不饱和聚酯，并且每双键分子量因子为154.4〜408作为分散相的微晶悬浮液和30〜97重量％的多羟基 官能度为2至约8，羟值为20至800，熔点低于30℃的化合物为粘结相适用于制备非细胞和细胞聚氨酯或聚氨酯基团， 含有聚异氰脲酸酯泡沫。

公开(公告)号：US07962594B2

公开(公告)日：2011-06-14

申请号：US11679158

申请日：2007-02-26

申请人： Stephane Kasriel ,  Walter Mann

发明人： Stephane Kasriel ,  Walter Mann

IPC分类号： G06F15/173 ,  G06F15/16

CPC分类号： H04L67/2814 ,  H04L67/2842 ,  H04L67/288

摘要： A system and method for delivering content on a network using differential caching is described. A request for information is received from a request-generating element coupled to a browser associated with a client through the network. A static portion and a dynamic portion of a document to be included in a response as the requested information are identified based on previously cached information by utilizing a software element that is not part of the client. The static portion is cached in a memory that is logically local to the client that requested the information. The static portion is served to the request-generating element from the cached memory. The dynamic portion is served to the request-generating element from the network. The static portion and the dynamic portion are integrated using the request-generating element into the document included in the response to the browser as the requested information.

摘要翻译： 描述了使用差分缓存在网络上传送内容的系统和方法。 从通过网络与与客户端相关联的浏览器耦合的请求生成元件接收到对信息的请求。 通过利用不是客户端的一部分的软件元素，基于先前缓存的信息来识别作为所请求的信息被包括在响应中的文档的静态部分和动态部分。 静态部分缓存在逻辑上位于请求信息的客户端的内存中。 静态部分从缓存的存储器提供给请求生成元素。 动态部分从网络提供给请求生成元素。 将静态部分和动态部分使用请求生成元素集成到包括在对浏览器的响应中的文档中作为所请求的信息。

公开(公告)号：US07647323B2

公开(公告)日：2010-01-12

申请号：US11221555

申请日：2005-09-08

申请人： Stephane Kasriel ,  Walter Mann ,  Sara Swanson

发明人： Stephane Kasriel ,  Walter Mann ,  Sara Swanson

IPC分类号： G06F17/30

CPC分类号： G06Q30/02 ,  G06F17/3089 ,  H04L67/24 ,  H04L69/329 ,  Y10S707/99945

摘要： A method of providing a web-site performance analysis service is disclosed that gathers data relative to visits to web-pages that are associated with a web-site. Traversals are identified that correspond to each visit to the web-pages. Each identified traversal has an identification of either an incoming link from which each visit originated or an outgoing link to which each visit terminated. A time duration is measured that corresponds to each visit. A chart of a statistical distribution of pages-per-visit of the identified traversals is formed. The identified traversals and time duration corresponding to each visit is stored in a memory. The stored data from the memory is transferred to a database. Performance measures are produced that correspond to each visit to the web-pages based on the stored data in the database. A web-site performance analysis system that implements this method also is disclosed.

摘要翻译： 公开了一种提供网站性能分析服务的方法，其收集与访问与网站相关联的网页的数据。 识别出与网页的每次访问相对应的遍历。 每个识别的遍历具有来自每个访问的入站链路的标识或每次访问终止的传出链路。 测量对应于每次访问的持续时间。 形成了所识别的遍历的每页访问的统计分布的图表。 所识别的遍历和对每次访问的持续时间存储在存储器中。 来自存储器的存储的数据被传送到数据库。 基于数据库中存储的数据，产生与网页的每次访问相对应的性能测量。 还公开了一种实现该方法的网站性能分析系统。