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

US20070074207A1 SPU task manager for cell processor 有权
标题翻译： SPU单元处理器任务管理器
公开(公告)号：US20070074207A1
公开(公告)日：2007-03-29
申请号：US11238087
申请日：2005-09-27
申请人： John Bates , Payton White , Richard Stenson , Howard Berkey , Atilla Vass , Mark Cerny , John Morgan
发明人： John Bates , Payton White , Richard Stenson , Howard Berkey , Atilla Vass , Mark Cerny , John Morgan
IPC分类号： G06F9/455
CPC分类号： G06F9/5027 , G06F9/4881 , G06F9/5033 , G06F2209/483 , G06F2212/251 , G06F2212/253
摘要： Cell processor task management in a cell processor having a main memory, one or more power processor units (PPU) and one or more synergistic processing units (SPU), each SPU having a processor and a local memory is described. An SPU task manager (STM) running on one or more of the SPUs reads one or more task definitions stored in the main memory into the local memory of a selected SPU. Based on information contained in the task definitions the SPU loads code and/or data related to the task definitions from the main memory into the local memory associated with the selected SPU. The selected SPU then performs one or more tasks using the code and/or data.
摘要翻译：描述具有主存储器，一个或多个功率处理器单元（PPU）和一个或多个协同处理单元（SPU）的单元处理器中的单元处理器任务管理，每个SPU具有处理器和本地存储器。在一个或多个SPU上运行的SPU任务管理器（STM）将存储在主存储器中的一个或多个任务定义读取到所选择的SPU的本地存储器中。基于任务定义中包含的信息，SPU将与主存储器相关的任务定义的代码和/或数据加载到与所选SPU相关联的本地存储器中。所选择的SPU然后使用代码和/或数据执行一个或多个任务。

2. 发明申请

US20050198290A1 Content distribution overlay network and methods for operating same in a P2P network 有权
标题翻译：内容分发覆盖网络和在P2P网络中操作相同的方法
公开(公告)号：US20050198290A1
公开(公告)日：2005-09-08
申请号：US10856426
申请日：2004-05-28
申请人： Howard Berkey , Payton White
发明人： Howard Berkey , Payton White
IPC分类号： G06F15/173 , H04L12/46 , H04L29/06 , H04L29/08
CPC分类号： H04L67/104 , H04L67/1002 , H04L67/1008 , H04L67/1046 , H04L67/1063 , H04L67/1068 , H04L67/108 , H04L67/1089 , H04L67/1093 , H04L67/34 , H04L69/329
摘要： A method, system and architecture for operating a content distribution overlay network in conjunction with a peer-to-peer network is provided. The method includes receiving a request for content at a content node from a requesting node of the peer-to-peer network. Then, referring to an overlay network to identify the content node as a head node of the overlay network. The head node has capabilities to transcode the content. The method then identifies a lowest level child node of the head node that has capabilities to transcode the content for the requesting node. The lowest level child node is then assigned to transcode the content for presentation to the requesting node, and each parent in the overlay network is assigned to transcode the content for its child. The lowest level child node receives the content transcoded from its parent in a format of the lowest level child node, and the overlay network is independent of the peer-to-peer network. In accordance with the method, nodes of the overlay network are arranged according to capability, and the head node has a highest capability of the overlay network.
摘要翻译：提供了一种与对等网络一起操作内容分发覆盖网络的方法，系统和架构。该方法包括从对等网络的请求节点在内容节点接收对内容的请求。然后，参考覆盖网络将内容节点标识为覆盖网络的头节点。头节点具有对内容进行转码的能力。该方法然后识别具有对请求节点的内容进行代码转换的能力的头节点的最低级子节点。然后分配最低级别的子节点对内容进行代码转换以呈现给请求节点，并且分配覆盖网络中的每个父代码对其子代码转码内容。最底层子节点以最低级子节点的格式接收从其父节点转码的内容，并且覆盖网络独立于对等网络。根据该方法，覆盖网络的节点根据能力进行布置，并且头节点具有覆盖网络的最高能力。

3. 发明申请

US20070121570A1 Broadcast messaging in peer to peer overlay network 有权
标题翻译：对等覆盖网络中的广播消息传递
公开(公告)号：US20070121570A1
公开(公告)日：2007-05-31
申请号：US11291121
申请日：2005-11-29
申请人： Yutaka Takeda , Howard Berkey , Payton White , Attila Vass
发明人： Yutaka Takeda , Howard Berkey , Payton White , Attila Vass
IPC分类号： H04B7/212
CPC分类号： H04L12/1854
摘要： Broadcast messages are efficiently directed to nodes of an overlay network. Broadcast messages include an End ID parameter specifying the range of key values for nodes that should receive the broadcast message. Each node of an overlay network maintains a list of finger nodes and their respective key values. Upon receiving a broadcast message, a node assigns a finger node a new End ID value based upon the End ID value of the broadcast message or the key value of an adjacent finger node. The node compares a finger node's new End ID value with the finger node's key value to determine whether to forward the broadcast message to that finger node. A broadcast message forwarded to a finger node includes an End ID parameter equal to the new End ID value determined for the finger node. Nodes can aggregate response messages from its finger nodes.
摘要翻译：广播消息被有效地指向覆盖网络的节点。广播消息包括指定应该接收广播消息的节点的密钥值的范围的结束ID参数。覆盖网络的每个节点维护手指节点及其各自的键值的列表。在接收到广播消息时，节点基于广播消息的结束ID值或相邻手指节点的密钥值来分配手指节点新的结束ID值。节点将手指节点的新结束ID值与手指节点的键值进行比较，以确定是否将广播消息转发到该手指节点。转发到手指节点的广播消息包括等于为手指节点确定的新的结束ID值的结束ID参数。节点可以聚合来自其手指节点的响应消息。

4. 发明授权

US08374086B2 Adaptive DHT node relay policies 有权
标题翻译：自适应DHT节点中继策略
公开(公告)号：US08374086B2
公开(公告)日：2013-02-12
申请号：US11759186
申请日：2007-06-06
申请人： James E. Marr , Payton White , Howard Berkey , Attila Vass
发明人： James E. Marr , Payton White , Howard Berkey , Attila Vass
IPC分类号： G01R31/08 , H04J1/16 , G06F15/16
CPC分类号： H04W40/22 , H04L45/745
摘要： In a distributed hash table overlay network, messages directed to a message ID are relayed to one or more nodes that have published interest in that message ID. Messages are distributed, processed, and/or cached in accordance with relay policies, which enable a wide variety of different applications. Nodes specify relay policies on a per-node, per-message, and/or per-message ID basis. Relay policies can specify any behavior associated with messages, including: message caching, distribution, and processing as well as migration of relay policies when nodes join or leave the network. Intermediate nodes can perform any arbitrary processing of messages based on relay policies. Intermediate nodes can offload processing to other nodes, either by forwarding messages to a different message ID or by enlisting the help of nodes subscribed to the message ID. Messages can include one or more subfields specifying commands and/or data to be evaluated by the intermediate node.
摘要翻译：在分布式散列表覆盖网络中，定向到消息ID的消息被中继到已经对该消息ID发表兴趣的一个或多个节点。消息根据中继策略进行分发，处理和/或高速缓存，从而能够实现各种不同的应用。节点在每节点，每消息和/或每消息ID的基础上指定中继策略。中继策略可以指定与消息相关联的任何行为，包括：消息缓存，分发和处理以及节点加入或离开网络时中继策略的迁移。中间节点可以根据中继策略对消息进行任意处理。中间节点可以通过将消息转发到不同的消息ID或通过登记订阅消息ID的节点的帮助来将处理卸载到其他节点。消息可以包括指定中间节点要评估的命令和/或数据的一个或多个子字段。

5. 发明申请

US20080304493A1 ADAPTIVE DHT NODE RELAY POLICIES 有权
标题翻译：自适应DHT节点中继策略
公开(公告)号：US20080304493A1
公开(公告)日：2008-12-11
申请号：US11759186
申请日：2007-06-06
申请人： James E. MARR , Payton White , Howard Berkey , Attila Vass
发明人： James E. MARR , Payton White , Howard Berkey , Attila Vass
IPC分类号： H04L12/56
CPC分类号： H04W40/22 , H04L45/745
摘要： In a distributed hash table overlay network, messages directed to a message ID are relayed one or more nodes that have published interest in that message ID. Messages are distributed, processed, and/or cached in accordance with relay policies, which enable a wide variety of different applications. Nodes to specify relay policies on a per-node, per-message, and/or per-message ID basis. Relay policies can specify any behavior associated with messages, including: message caching, distribution, and processing as well as migration of relay policies when nodes join or leave the network. Intermediate nodes can perform any arbitrary processing of message based on relay policies. Intermediate nodes can offload processing to other nodes, either by forwarding messages to a different message ID or by enlisting the help of nodes subscribed to the message ID. Messages can include one or more subfields specifying commands and/or data to be evaluated by the intermediate node.
摘要翻译：在分布式哈希表覆盖网络中，定向到消息ID的消息被中继到已经对该消息ID发表兴趣的一个或多个节点。消息根据中继策略进行分发，处理和/或高速缓存，从而能够实现各种不同的应用。节点在每个节点，每个消息和/或每个消息ID的基础上指定中继策略。中继策略可以指定与消息相关联的任何行为，包括：消息缓存，分发和处理以及节点加入或离开网络时中继策略的迁移。中间节点可以根据中继策略对消息进行任意处理。中间节点可以通过将消息转发到不同的消息ID或通过登记订阅消息ID的节点的帮助来将处理卸载到其他节点。消息可以包括指定中间节点要评估的命令和/或数据的一个或多个子字段。

6. 发明申请

US20070016572A1 Predictive user interface 有权
标题翻译：预测用户界面
公开(公告)号：US20070016572A1
公开(公告)日：2007-01-18
申请号：US11181540
申请日：2005-07-13
申请人： John Bates , Payton White , Yutaka Takeda
发明人： John Bates , Payton White , Yutaka Takeda
IPC分类号： G06F17/30
CPC分类号： G06F3/0236 , G06F3/0237 , G06F3/04817 , Y10S707/99935
摘要： A user interface enables the input of text and other complicated inputs by identifying entities having an increased likelihood of being selected and decreasing the precision needed to select these entities. The user interface displays a set of entities in a radial pattern. The user manipulates a pointer with a joystick or other input device to select entities. The entities having a higher selection probabilities are allocated more space in the pattern than the other entities. The entities having a higher selection probabilities may also be emphasized with visual cues. Selection probabilities may be determined by comparing a sequence of entities previously selected, such as a partially input word, with a set of commonly selected sequences of entities, such as a set of words commonly used by users. The user interface can display a list of words corresponding with a sequence of selected entities, including characters from different writing systems.
摘要翻译：用户接口通过识别具有增加的可能性被选择的实体并且降低选择这些实体所需的精度来实现文本和其他复杂输入的输入。用户界面以径向模式显示一组实体。用户使用操纵杆或其他输入设备操纵指针以选择实体。具有较高选择概率的实体在模式中被分配比其他实体更多的空间。具有更高选择概率的实体也可以用视觉线索强调。可以通过将先前选择的实体的序列（例如部分输入的字）与一组常用的实体序列（诸如用户通常使用的一组单词）进行比较来确定选择概率。用户界面可以显示与选定实体的序列相对应的单词列表，包括来自不同书写系统的字符。

7. 发明申请

US20070074212A1 Cell processor methods and apparatus 有权
标题翻译：单元处理器的方法和装置
公开(公告)号：US20070074212A1
公开(公告)日：2007-03-29
申请号：US11238077
申请日：2005-09-27
申请人： John Bates , Payton White , Attila Vass
发明人： John Bates , Payton White , Attila Vass
IPC分类号： G06F9/46
CPC分类号： G06F9/5066 , G06F9/5083
摘要： Methods and apparatus for cell processors are disclosed. A policy module is loaded from a main memory of a cell processor into the local memory of a selected synergistic processing unit (SPU) under control of an SPU policy module manager (SPMM) running on the SPU. The policy module loads a work queue from the main memory into the local memory of the SPU. The policy module interprets and process one or more tasks from the work queue on the SPU. The selected SPU performs the task(s) and after completion or upon a pre-emption, returns control of the SPU to the SPMM.
摘要翻译：公开了用于单元处理器的方法和装置。在SPU上运行的SPU策略模块管理器（SPMM）的控制下，将策略模块从小区处理器的主存储器加载到所选择的协同处理单元（SPU）的本地存储器中。策略模块将工作队列从主存储器加载到SPU的本地存储器中。策略模块从SPU上的工作队列中解释和处理一个或多个任务。选定的SPU执行任务，完成后或优先选择后，将SPU的控制权返回给SPMM。

8. 发明申请

US20050240937A1 Specifying parameters for selective return to an invoker 失效
标题翻译：指定参数以选择性地返回到调用者
公开(公告)号：US20050240937A1
公开(公告)日：2005-10-27
申请号：US10832917
申请日：2004-04-26
申请人： John Bates , Payton White
发明人： John Bates , Payton White
IPC分类号： G06F9/46 , G06F3/00
CPC分类号： G06F9/548 , G06F9/465 , G06F9/547
摘要： Provided is a method and an apparatus for selectively returning parameters from an invoked object. Specifically, a requestor can invoke an object by transmitting a message. The message can include specified return parameters from an interface, such as an interface definition language (IDL). Alternatively, an object stub locally accessible by the requester can be invoked. The object stub, similar to the IDL, can specify the parameters to return. After the execution of a method in the invoked object, the requestor receives the return parameters or a subset of the return parameters previously selected.
摘要翻译：提供了一种用于从被调用对象中选择性地返回参数的方法和装置。具体来说，请求者可以通过发送消息来调用对象。消息可以包括来自接口的指定的返回参数，例如接口定义语言（IDL）。或者，可以调用可由请求者本地访问的对象存根。对象存根类似于IDL，可以指定要返回的参数。在调用对象中执行方法之后，请求者接收返回参数或先前选择的返回参数的子集。

9. 发明申请

US20050021758A1 Method and system for identifying available resources in a peer-to-peer network 有权
标题翻译：用于识别对等网络中的可用资源的方法和系统
公开(公告)号：US20050021758A1
公开(公告)日：2005-01-27
申请号：US10857559
申请日：2004-05-27
申请人： Payton White
发明人： Payton White
IPC分类号： G06F15/173 , H04L29/06 , H04L29/08
CPC分类号： H04L67/16 , H04L67/104 , H04L67/1068 , H04L67/1089
摘要： A method for a peer of a peer-to-peer network to search for an available resource is provided. The method initiates with representing a request for the available resource as a request bit string. The method includes transmitting the request to a parent of the peer. Then, an aggregate availability bit string associated with the parent is accessed. The aggregate availability bit string represents an aggregate resource availability for all children of the parent. Next, a logical operation is performed with the bits of the aggregate availability bit string and the bits of the request bit string. Then, it is determined whether any of the children of the parent are able to provide the available resource based upon a result of the logical operation. A peer-to-peer grid and a system for publishing and identifying an available resource in a peer-to-peer distributed network are also provided.
摘要翻译：提供了一种用于对等网络的对等体来搜索可用资源的方法。该方法通过将可用资源的请求表示为请求位串来发起。该方法包括将请求发送到对等体的父节点。然后，访问与父级关联的聚合可用性位串。聚合可用性位串表示父级的所有子级的总体资源可用性。接下来，利用聚合可用性比特串的比特和请求比特串的比特来执行逻辑运算。然后，基于逻辑运算的结果确定父节点的任何子节点是否能够提供可用资源。还提供了点对点网格和用于在对等分布式网络中发布和识别可用资源的系统。

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