专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20070118652A1 Bundling and sending work units to a server based on a weighted cost 有权
标题翻译：根据加权成本将工作单元捆绑并发送到服务器
公开(公告)号：US20070118652A1
公开(公告)日：2007-05-24
申请号：US11262129
申请日：2005-10-28
申请人： Richard Diedrich , Jinmei Shen , Hao Wang
发明人： Richard Diedrich , Jinmei Shen , Hao Wang
IPC分类号： G06F15/173
CPC分类号： G06F9/5027 , H04L67/1002 , H04L67/1008 , H04L67/1012
摘要： A method, apparatus, system, and signal-bearing medium that, in an embodiment, accumulate multiple work units into a bundle and send the bundle to a selected active server. The active server is selected based on the relative costs for the work units and costs for servers. In an embodiment, selecting the active server includes determining relative work unit costs and relative server costs, calculating total weighted costs for the servers based on the relative work unit costs and the relative server costs, and selecting the active server that has the lowest total weighted cost. The active server may either process a work in the bundle locally or send the work unit to a passive server for processing. The active server determines whether each work unit can be executed at the active server. If the work unit can be executed on the active server, the active server executes the work unit and creates local results. If the work unit cannot be executed on the server, the active server sends the work unit to a passive server, and the passive server creates remote results. The active server receives the remote results from the passive server, combines the local results with the remote results into combined results, and sends the combined results to a client. In various embodiments, branch and dependency data are included in the bundle, which the active server uses to order and conditionally process for the work units.
摘要翻译：一种方法，装置，系统和信号承载介质，其在一个实施例中将多个工作单元累积成束并将捆绑包发送到所选择的活动服务器。基于工作单位的相对成本和服务器的成本来选择活动服务器。在一个实施例中，选择活动服务器包括确定相对工作单元成本和相对服务器成本，基于相对工作单元成本和相对服务器成本来计算服务器的总加权成本，以及选择具有最低总加权的活动服务器成本。活动服务器可以在本地处理捆绑中的工作，也可以将工作单元发送到被动服务器进行处理。活动服务器确定是否可以在活动服务器上执行每个工作单元。如果可以在活动服务器上执行工作单元，则活动服务器将执行工作单元并创建本地结果。如果无法在服务器上执行工作单元，则主动服务器将工作单元发送到被动服务器，被动服务器创建远程结果。活动服务器从被动服务器接收远程结果，将本地结果与远程结果相结合，并将结果发送给客户端。在各种实施例中，分支和依赖关系数据被包括在分组中，活动服务器使用该数据包对工作单元进行订购和有条件地处理。

2. 发明申请

US20070083521A1 Routing requests based on synchronization levels 审中-公开
标题翻译：基于同步级别路由请求
公开(公告)号：US20070083521A1
公开(公告)日：2007-04-12
申请号：US11246821
申请日：2005-10-07
申请人： Richard Diedrich , Jinmei Shen , Hao Wang
发明人： Richard Diedrich , Jinmei Shen , Hao Wang
IPC分类号： G06F17/30
CPC分类号： H04L67/1095 , H04L67/327
摘要： A method, apparatus, system, and signal-bearing medium that, in an embodiment, route requests to servers based on a synchronization level of data that the servers provide. In an embodiment, synchronization levels that servers provide are determined, a synchronization level that a request requires is determined, a server is selected based on the provided synchronization levels and the required synchronization level, and the request is routed to the selected server. The selection of the server may include selecting a subset of the servers, ordering the subset based on the provided synchronization levels, and selecting the highest synchronization level that is processing less than a threshold number of requests. In various embodiments, the provided synchronization levels are determined based on probabilities that data changes are synchronized between the servers based on distributions of propagation time delays of data changes between the servers, based on distributions of elapsed times between data changes, and based on both distributions.
摘要翻译：一种方法，装置，系统和信号承载介质，其在一个实施例中基于服务器提供的数据的同步级别将请求路由到服务器。在一个实施例中，确定服务器提供的同步级别，确定请求所需的同步级别，基于所提供的同步级别和所需的同步级别来选择服务器，并且该请求被路由到所选择的服务器。服务器的选择可以包括选择服务器的子集，基于所提供的同步级别对子集进行排序，以及选择处理小于阈值数目的请求的最高同步级别。在各种实施例中，所提供的同步水平是基于基于数据变化之间的数据变化的传播时间延迟的分布，基于数据变化之间的经过时间的分布，并且基于两个分布。

3. 发明申请

US20050197787A1 Use of dominance to improve performance or increase search space in genetic algorithms 有权
标题翻译：利用优势来提高遗传算法的性能或增加搜索空间
公开(公告)号：US20050197787A1
公开(公告)日：2005-09-08
申请号：US10793527
申请日：2004-03-04
申请人： Michael Diedrich , Richard Diedrich
发明人： Michael Diedrich , Richard Diedrich
IPC分类号： G06F17/00 , G06F17/20
CPC分类号： G06N3/126
摘要： Methods, systems, and articles of manufacture utilizing the biological principle of dominance to affect the operation of a genetic algorithm are provided. The genetic algorithm may be used to select a multi-parameter solution set to a user-defined problem, with each parameter modeled as a gene. Dominance may be introduced by associating a dominance attribute with one or more gene values (i.e., traits), for example, that are known to lead to good solution sets, allowing the corresponding gene values to be specified as dominant or recessive. By specifying the known good gene values as dominant, system performance may be optimized as the genetic algorithm may quickly arrive at a good solution set (e.g., due to the known good value appearing more often in future generations). On the other hand, by specifying the known good gene value as recessive, the search space may be increased as the genetic algorithm may evaluate a larger number of possible solution sets (e.g., due to the known good value appearing less often in future generations).
摘要翻译：提供了利用优势生物学原理影响遗传算法运行的方法，系统和制品。遗传算法可用于选择用户定义问题的多参数解，将每个参数建模为基因。可以通过将优势属性与一个或多个基因值（即，特征）相关联来引入优势，例如，已知其导致良好的溶液组，允许相应的基因值被指定为显性或隐性。通过将已知的良好基因值指定为主导，可以优化系统性能，因为遗传算法可以快速达到良好的解集（例如，由于后代中更常见的已知良好值）。另一方面，通过将已知的良好基因值指定为隐性，可以增加搜索空间，因为遗传算法可以评估更多数量的可能解集（例如，由于在后代中较少出现的已知好价值）。

4. 发明申请

US20070083642A1 Fully distributed data collection and consumption to maximize the usage of context, resource, and capacity-based client server interactions 审中-公开
标题翻译：全面分布的数据收集和消费，以最大限度地利用上下文，资源和基于能力的客户端服务器交互
公开(公告)号：US20070083642A1
公开(公告)日：2007-04-12
申请号：US11246822
申请日：2005-10-07
申请人： Richard Diedrich , Jinmei Shen , Hao Wang
发明人： Richard Diedrich , Jinmei Shen , Hao Wang
IPC分类号： G06F15/173
CPC分类号： H04L67/327 , H04L41/0233 , H04L41/0896 , H04L43/00 , H04L43/0817
摘要： A method, distributed-computing system, and computer program product for providing efficient workload management within a distributed computing environment. Each device within the distributed-computing environment is enhanced with a workload management controller (WLMC) functionality/utility, designed specifically for the type of device (i.e., client WLMC versus server WLMC) and utilized to collect process data about the particular device (e.g., status information) and about the device's interaction with the network. With the localized device-based WLM Controllers, each device utilizes fully distributed tagged information to accomplish capacity-based routing, context-based routing, and resource-based routing without any overhead or loss of data and without any network congestion. The distributed WLM Controller model enables each device to operate without concern for the level of CPU usage or memory usage of the particular device.
摘要翻译：一种方法，分布式计算系统和计算机程序产品，用于在分布式计算环境中提供高效的工作负载管理。分布式计算环境中的每个设备都通过专门针对设备类型（即客户端WLMC与服务器WLMC）专门设计的工作负载管理控制器（WLMC）功能/实用程序进行了增强，并用于收集有关特定设备的过程数据（例如，，状态信息）以及设备与网络的交互。通过本地化的基于设备的WLM控制器，每个设备都使用完全分布式的标记信息来完成基于容量的路由，基于上下文的路由和基于资源的路由，而不会造成数据的任何开销或丢失，并且没有任何网络拥塞。分布式WLM控制器模型使每个设备能够在不考虑特定设备的CPU使用水平或内存使用情况的情况下运行。

5. 发明申请

US20080022269A1 Method and Apparatus for Building Executable Computer Programs Using Compiled Program Libraries 失效
标题翻译：使用编译程序库构建可执行计算机程序的方法和装置
公开(公告)号：US20080022269A1
公开(公告)日：2008-01-24
申请号：US11835601
申请日：2007-08-08
申请人： Richard Diedrich , Richard Saltness , John Santosuosso
发明人： Richard Diedrich , Richard Saltness , John Santosuosso
IPC分类号： G06F9/45
CPC分类号： G06F8/54
摘要： A program which is linked or bound by reference (referenced program) into one or more other programs supports multiple valid export signatures, each corresponding to a respective version of the referenced program. When a program is built, it records the current signature of each referenced program it is bound to. When subsequently determining whether to rebuild the program, the previously recorded signature of any referenced program is compared with all currently supported signatures of the referenced program, and only if none of the supported signatures matches does the program need to be re-built to ensure compatibility with the referenced program. Preferably, the referenced program is a program library containing supporting procedures, intended to be used by multiple applications programs which are bound to it.
摘要翻译：通过引用（引用程序）链接或绑定到一个或多个其他程序中的程序支持多个有效的导出签名，每个对应于引用程序的相应版本。当一个程序被构建时，它会记录它所绑定的每个被引用程序的当前签名。当随后确定是否重建程序时，将所引用程序的先前记录的签名与被引用程序的所有当前支持的签名进行比较，并且只有当所支持的签名中没有一个匹配时才需要重新构建该程序以确保兼容性与参考程序。优选地，所引用的程序是包含支持程序的程序库，其旨在被绑定到其上的多个应用程序使用。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式