公开(公告)号：US5913041A

公开(公告)日：1999-06-15

申请号：US762355

申请日：1996-12-09

申请人： Srinivas Ramanathan ,  Tai Jin ,  Manjari Asawa

发明人： Srinivas Ramanathan ,  Tai Jin ,  Manjari Asawa

IPC分类号： H04L12/26 ,  H04L29/06 ,  H04L29/08 ,  G06F13/00

CPC分类号： H04L43/00 ,  H04L12/2602 ,  H04L29/06 ,  H04L63/0245 ,  H04L43/0829 ,  H04L43/0864 ,  H04L43/0888 ,  H04L43/106 ,  H04L43/16 ,  H04L67/00 ,  H04L69/329

摘要： A method and system of monitoring throughput of a data access system includes logging each transfer of data from a content server to a remote site, with each log entry including information indicative of transfer size, date, times, source and destination. The method includes accessing the log information in a passive and non-intrusive manner to evaluate the performance of transfers to a selected subset of the remote sites. In another embodiment, the performance evaluation is implemented for system resource allocation planning. In the preferred embodiment, the data access system is a broadband data system and the content servers utilize Internet applications. Also in the preferred embodiment, the data throughput is measured by the transfer rate of useful data, rather than all data including retransmissions.

摘要翻译： 监视数据访问系统的吞吐量的方法和系统包括记录从内容服务器到远程站点的数据的每次传送，每个日志条目包括指示传送大小，日期，时间，源和目的地的信息。 该方法包括以被动和非侵入的方式访问日志信息，以评估到远程站点的选定子集的传输性能。 在另一个实施例中，对系统资源分配规划进行性能评估。 在优选实施例中，数据访问系统是宽带数据系统，内容服务器利用因特网应用。 同样在优选实施例中，数据吞吐量通过有用数据的传输速率来测量，而不是包括重传的所有数据。

公开(公告)号：US6041041A

公开(公告)日：2000-03-21

申请号：US838117

申请日：1997-04-15

申请人： Srinivas Ramanathan ,  Edward H. Perry ,  Tai Jin

发明人： Srinivas Ramanathan ,  Edward H. Perry ,  Tai Jin

IPC分类号： G06F13/00 ,  H04L12/24 ,  H04L29/14 ,  H04J3/14

CPC分类号： H04L41/0213

摘要： A scheme is described for a data service system having a number of modules. Some of the modules are interdependent. To measure the status of an individual module, the scheme first collects measurements from a number of measurement routes that involve the module. Then the scheme analyzes the interdependencies of the measurements to determine the status of the individual module. The scheme may also determine status of the data service system with a minimal number of measurement routes. This is done by determining (1) all possible measurement routes, (2) determining the dependency between the modules and the measurement routes, and (3) analyzing the dependency to select minimal number of the measurement routes. The scheme can diagnose whether a module is a problematic module or not by analyzing a number of measurements that involve the module. If one of the measurements is good, the module is identified as non-problematic. The data service system may also include a test target coupled to a networking module of the data service system to allow service test signals to be measured through the network module.

摘要翻译： 对具有多个模块的数据服务系统描述了一种方案。 一些模块是相互依赖的。 为了测量单个模块的状态，该方案首先从涉及模块的多个测量路径收集测量值。 然后，该方案分析测量的相互依赖性，以确定单个模块的状态。 该方案还可以以最小数量的测量路由确定数据服务系统的状态。 这通过确定（1）所有可能的测量路线，（2）确定模块与测量路线之间的依赖关系，以及（3）分析依赖关系以选择最小数量的测量路线来完成。 该方案可以通过分析涉及该模块的多个测量来诊断模块是否是有问题的模块。 如果其中一个测量值很好，则该模块被识别为无问题。 数据服务系统还可以包括耦合到数据服务系统的网络模块的测试目标，以允许通过网络模块测量服务测试信号。

公开(公告)号：US06425057B1

公开(公告)日：2002-07-23

申请号：US09140977

申请日：1998-08-27

申请人： Ludmila Cherkasova ,  Martin F. Arlitt ,  Richard J. Friedrich ,  Tai Jin

发明人： Ludmila Cherkasova ,  Martin F. Arlitt ,  Richard J. Friedrich ,  Tai Jin

IPC分类号： G06F1212

CPC分类号： G06F17/30607 ,  G06F12/121 ,  G06F12/122

摘要： A method and system for caching objects and replacing cached objects in an object-transfer environment maintain a dynamic indicator (Pr(f)) for each cached object, with the dynamic indicator being responsive to the frequency of requests for the object and being indicative of the time of storing the cached object relative to storing other cached objects. In a preferred embodiment, the size of the object is also a factor in determining the dynamic indicator of the object. In the most preferred embodiment, the cost of obtaining the object is also a factor. A count of the frequency of requests and the use of the relative time of storage counterbalance each other with respect to maintaining a cached object in local cache. That is, a high frequency of requests favors maintaining the object in cache, but a long period of cache favors evicting the object. Thus, cache pollution is less likely to occur.

摘要翻译： 用于缓存对象并替换对象传送环境中的缓存对象的方法和系统为每个缓存对象维护动态指示符（Pr（f）），动态指示符响应于对象的请求频率， 存储缓存对象相对于存储其他缓存对象的时间。 在优选实施例中，对象的大小也是确定对象的动态指示符的因素。 在最优选的实施例中，获得对象的成本也是一个因素。 关于在本地高速缓存中维护缓存的对象，请求的频率的计数和相对于存储的相对时间的使用相互平衡。 也就是说，高频率的请求有利于将对象保持在缓存中，但长时间的缓存有利于驱逐对象。 因此，缓存污染不太可能发生。

公开(公告)号：US06304906B1

公开(公告)日：2001-10-16

申请号：US09130636

申请日：1998-08-06

申请人： Nina T. Bhatti ,  Richard J. Friedrich ,  Tai Jin ,  Anna M. Zara

发明人： Nina T. Bhatti ,  Richard J. Friedrich ,  Tai Jin ,  Anna M. Zara

IPC分类号： G06F1516

CPC分类号： H04L12/2856 ,  H04L12/2874 ,  H04L47/10 ,  H04L47/15 ,  H04L47/19 ,  H04L47/24 ,  H04L47/70 ,  H04L47/808 ,  H04L47/821 ,  H04L63/102 ,  H04L67/02 ,  H04L67/327

摘要： A data service system is accessible through a plurality of access requests. The data service system includes a user access request classification system for classifying the access requests. The user access request classification system includes a request classifier that classifies each of the access requests into one of a plurality of classes based on a predetermined classification policy. The user access request classification system also includes a plurality of buffers coupled to the request classifier. Each buffer corresponds to one of the classes to receive the access requests classified by the request classifier as belonging to that class such that the data service system processes the access requests in accordance with their classes. A method of classifying user access requests for the data service system is also described.

摘要翻译： 数据服务系统可通过多个访问请求访问。 数据服务系统包括用于对访问请求进行分类的用户访问请求分类系统。 用户访问请求分类系统包括：请求分类器，其基于预定的分类策略将每个访问请求分类为多个类别中的一个。 用户访问请求分类系统还包括耦合到请求分类器的多个缓冲器。 每个缓冲器对应于接收由请求分类器归类为属于该类的访问请求的类之一，使得数据服务系统根据其类来处理访问请求。 还描述了对数据服务系统的用户访问请求进行分类的方法。

公开(公告)号：US06189033B1

公开(公告)日：2001-02-13

申请号：US09116716

申请日：1998-07-16

申请人： Tai Jin ,  James D. Salehi ,  Richard J. Friedrich ,  Nina T. Bhatti

发明人： Tai Jin ,  James D. Salehi ,  Richard J. Friedrich ,  Nina T. Bhatti

IPC分类号： G06F15173

CPC分类号： G06F17/30902 ,  H04L29/12009 ,  H04L29/12207 ,  H04L61/20 ,  H04L67/1002 ,  H04L67/1008 ,  H04L67/1012 ,  H04L67/1014 ,  H04L67/322

摘要： A data service system is described that includes a plurality of data content sites and is coupled to a number of user terminals via a network. The data service system also includes a system for providing performance guarantees for the content sites. The system for providing performance guarantees includes a plurality of class processors, each for controlling access throughput of one of the content sites in accordance with a predetermined admission control policy. When one of the class processors receives an access request for one of the content sites, that class processor determines if the access request should be accepted based on the corresponding admission control policy and sends the access request to the corresponding content site when accepting the access request. The system also includes a classifier that determines to which one of the class processors the access request should be sent. A method of controlling access throughput of the content sites is also described.

摘要翻译： 描述了包括多个数据内容站点并且经由网络耦合到多个用户终端的数据服务系统。 数据服务系统还包括用于为内容站点提供性能保证的系统。 用于提供性能保证的系统包括多个类处理器，每个处理器用于根据预定的准入控制策略来控制一个内容站点的访问吞吐量。 当其中一个类处理器接收到一个内容站点的访问请求时，该类处理器根据对应的准入控制策略确定是否应该接受访问请求，并且在接受访问请求时将访问请求发送到对应的内容站点 。 该系统还包括分类器，其确定应该向哪个类处理器发送访问请求。 还描述了一种控制内容站点的访问吞吐量的方法。