-
1.发明授权System for learning and applying integrated task and data parallel strategies in dynamic applications 失效在动态应用中学习和应用综合任务和数据并行策略的系统公开(公告)号：US06675189B2
公开(公告)日：2004-01-06
申请号：US09086032
申请日：1998-05-28
IPC分类号： G06F900
CPC分类号： G06N99/005
摘要： A system for learning and applying a task and data parallel strategy to an application that includes at least one task for processing an input data stream to produce an output data stream includes the following components. A controller measuring an execution of the application to generate an action space representing a task and data parallel strategy. A run-time system applying the action space to implement the task and data parallel strategy.
摘要翻译： 一种用于向包括用于处理输入数据流以产生输出数据流的至少一个任务的应用的学习应用任务和数据并行策略的系统包括以下组件。 测量应用程序执行以生成表示任务和数据并行策略的动作空间的控制器。 运行时系统应用动作空间来实现任务和数据并行策略。
-
2.发明授权公开(公告)号：US06480876B2
公开(公告)日：2002-11-12
申请号：US09085795
申请日：1998-05-28
IPC分类号： G06F900
CPC分类号： G06F9/5066
摘要： A system for integrating task and data parallelism in a dynamic application that includes at least one task for processing an input data stream to produce an output data stream replaces the at least one task with the following components. A splitter task for partitioning the input data stream into a plurality of data chunks. A plurality of worker tasks for processing subsets of the data chunks, each worker task being an instance of the at least one task, and a joiner task combining the processed data chunks to produce the output data stream.
摘要翻译： 用于在动态应用程序中集成任务和数据并行性的系统，其包括用于处理输入数据流以产生输出数据流的至少一个任务，用下列组件代替至少一个任务。 用于将输入数据流分割成多个数据块的分离器任务。 用于处理数据块的子集的多个工作任务，每个工作任务是至少一个任务的实例，以及组合所处理的数据块以产生输出数据流的加入任务。
-
公开(公告)号：US6067604A
公开(公告)日：2000-05-23
申请号：US909405
申请日：1997-08-11
申请人： Umakishore Ramachandran , Robert H. Halstead, Jr. , Christopher F. Joerg , Leonidas Kontothanassis , Rishiyur S. Nikhil , James M. Rehg
发明人： Umakishore Ramachandran , Robert H. Halstead, Jr. , Christopher F. Joerg , Leonidas Kontothanassis , Rishiyur S. Nikhil , James M. Rehg
CPC分类号： G06F9/544 , G06F9/52 , Y10S707/99953 , Y10S707/99957
摘要： In a computer system, a memory is allocated to a plurality of ports. The ports are arranged in a spatial ordering. A plurality of various sized data items are temporally ordered in each of the plurality of ports. Each data item includes a time-stamp to indicate the temporal ordering of the plurality of data items. The plurality of data items are atomically accessed by a plurality of threads using space and time coordinates. The space and time coordinates uniquely identify each of the plurality of data items.
摘要翻译： 在计算机系统中，向多个端口分配存储器。 端口按空间排列排列。 多个不同大小的数据项在多个端口的每一个中被暂时排列。 每个数据项包括指示多个数据项的时间顺序的时间戳。 使用空间和时间坐标由多个线程原子地访问多个数据项。 空间和时间坐标唯一地标识多个数据项中的每一个。
-
公开(公告)号：US07139784B2
公开(公告)日：2006-11-21
申请号：US10629357
申请日：2003-07-29
IPC分类号： G06F17/30
CPC分类号： G06F12/0269 , Y10S707/99957
摘要： The performance of an application is improved by identifying and eliminating items with dead time-stamps and eliminating work on items with irrelevant time-stamps. An algorithm executing in each node of a task graph computes and propagates guarantees which are used to eliminate both items with dead time-stamps and irrelevant computation on dead time-stamps. A continuous garbage collector eliminates items with dead time-stamps while the node continues to process received items. Unnecessary computations are reduced by automatically discerning the interest set of downstream modules for time-stamps and feeding the interest set back to upstream modules.
摘要翻译： 通过识别和消除带有死时戳的物品，消除对不合时宜的物品的工作，可以提高应用程序的性能。 在任务图的每个节点中执行的算法计算和传播保证，其用于消除具有死时戳的两个项目以及对死时间戳的不相关的计算。 连续的垃圾收集器消除了具有死区时间戳的项目，同时节点继续处理收到的项目。 通过自动识别下游模块对时间戳的兴趣集，将兴趣集合回馈给上游模块，可以减少不必要的计算。
-
公开(公告)号：US20050027760A1
公开(公告)日：2005-02-03
申请号：US10629357
申请日：2003-07-29
CPC分类号： G06F12/0269 , Y10S707/99957
摘要： The performance of an application is improved by identifying and eliminating items with dead time-stamps and eliminating work on items with irrelevant time-stamps. An algorithm executing in each node of a task graph computes and propagates guarantees which are used to eliminate both items with dead time-stamps and irrelevant computation on dead time-stamps. A continuous garbage collector eliminates items with dead time-stamps while the node continues to process received items. Unnecessary computations are reduced by automatically discerning the interest set of downstream modules for time-stamps and feeding the interest set back to upstream modules.
摘要翻译： 通过识别和消除带有死时戳的物品，消除对不合时宜的物品的工作，可以提高应用程序的性能。 在任务图的每个节点中执行的算法计算和传播保证，其用于消除具有死时戳的两个项目以及对死时间戳的不相关的计算。 连续的垃圾收集器消除了具有死区时间戳的项目，同时节点继续处理收到的项目。 通过自动识别下游模块对时间戳的兴趣集，将兴趣集合回馈给上游模块，可以减少不必要的计算。
-
公开(公告)号：US20080002740A1
公开(公告)日：2008-01-03
申请号：US11771221
申请日：2007-06-29
IPC分类号： H04J3/16
CPC分类号： H04L69/32
摘要： An exemplary device comprises a first and second protocol layer, and an information exchange service (IES) in communication with the first and the second protocol layers. The IES is configured to share control information between the protocol layers. A protocol layer is capable of adapting behavior based on the control information shared by the IES. An exemplary method is presented for communicating between publishing and subscribing layers in a protocol stack. The method comprises, in the publishing layer, specifying a selected attribute and an associated value to be shared with the subscribing layer. The method further comprises storing the specified attribute and associated value, and receiving a subscribe request from the requesting layer. The request specifies the published attribute. The method further comprises, in response to the request, providing the associated value to a subscribing layer, and in the subscribing layer, adapting behavior based on the shared control information.
摘要翻译： 示例性设备包括第一和第二协议层以及与第一和第二协议层通信的信息交换服务（IES）。 IES被配置为在协议层之间共享控制信息。 协议层能够根据IES共享的控制信息调整行为。 呈现了用于在协议栈中的发布和订阅层之间进行通信的示例性方法。 该方法包括在发布层中指定要与订阅层共享的所选择的属性和相关联的值。 该方法还包括存储指定的属性和相关联的值，以及从请求层接收订阅请求。 该请求指定已发布的属性。 该方法还包括响应于该请求，向订阅层提供相关联的值，并且在订阅层中，基于共享控制信息调整行为。
-
公开(公告)号：US07693181B2
公开(公告)日：2010-04-06
申请号：US11771221
申请日：2007-06-29
IPC分类号： H04J3/16
CPC分类号： H04L69/32
摘要： An exemplary device comprises a first and second protocol layer, and an information exchange service (IES) in communication with the first and the second protocol layers. The IES is configured to share control information between the protocol layers. A protocol layer is capable of adapting behavior based on the control information shared by the IES. An exemplary method is presented for communicating between publishing and subscribing layers in a protocol stack. The method comprises, in the publishing layer, specifying a selected attribute and an associated value to be shared with the subscribing layer. The method further comprises storing the specified attribute and associated value, and receiving a subscribe request from the requesting layer. The request specifies the published attribute. The method further comprises, in response to the request, providing the associated value to a subscribing layer, and in the subscribing layer, adapting behavior based on the shared control information.
摘要翻译： 示例性设备包括第一和第二协议层以及与第一和第二协议层通信的信息交换服务（IES）。 IES被配置为在协议层之间共享控制信息。 协议层能够根据IES共享的控制信息调整行为。 呈现了用于在协议栈中的发布和订阅层之间进行通信的示例性方法。 该方法包括在发布层中指定要与订阅层共享的所选择的属性和相关联的值。 该方法还包括存储指定的属性和相关联的值，以及从请求层接收订阅请求。 该请求指定已发布的属性。 该方法还包括响应于该请求，向订阅层提供相关联的值，并且在订阅层中，基于共享控制信息调整行为。
-
-
-
-
-
-
搜索结果
7 条记录 (耗时 0.022 秒)
