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. 发明公开

EP2946298A1 CONTEXT SWITCHING WITH OFFLOAD PROCESSORS 无效
标题翻译：具有输出PROCESSORS上下文切换
公开(公告)号：EP2946298A1
公开(公告)日：2015-11-25
申请号：EP13872212.9
申请日：2013-06-26
申请人： Xockets IP LLC
发明人： DALAL, Parin , BELAIR, Stephen
IPC分类号： G06F13/00
CPC分类号： G06F13/1652 , G06F9/3877 , G06F9/461 , G06F9/4843 , G06F12/0815 , G06F12/0875 , G06F13/16 , G06F13/285 , G06F13/362 , G06F13/4022 , G06F13/4068 , G06F15/161 , G06F15/17337 , H04L29/08135 , H04L29/08549 , H04L47/193 , H04L47/2441 , H04L47/56 , H04L47/624 , H04L47/6295 , H04L49/40 , H04L49/90 , H04L61/103 , H04L61/2592 , H04L61/6086 , Y02D10/14 , Y02D10/151
摘要： A method for context switching of multiple offload processors is disclosed. The method can include receiving network packets for processing through a memory bus connected socket, organizing the network packets into multiple sessions for processing, suspending processing of at least one session by reading a cache state of at least one of the offload processor into a context memory by operation of a scheduling circuit, with virtual memory locations and physical cache locations being aligned, and subsequently directing transfer of the cache state to at least one of the offload processors for processing by operation of the scheduling circuit.

2. 发明公开

EP1829328A1 SYSTEM AND METHOD FOR SCALABLE DATA DISTRIBUTION 有权转让
标题翻译：系统和方法可扩展的数据分布
公开(公告)号：EP1829328A1
公开(公告)日：2007-09-05
申请号：EP05849756.1
申请日：2005-12-09
申请人： Google, Inc.
发明人： JAIN, Arvind
IPC分类号： H04L29/06
CPC分类号： H04L29/08135 , H04L67/10 , H04L67/1002 , H04L67/1008 , H04L67/1021 , H04L67/1029 , H04L67/1034 , H04L67/1095 , H04L67/16
摘要： A system having a resource manager, a plurality of masters, and a plurality of slaves, interconnected by a communications network. To distribute data, a master determined that a destination slave of the plurality slaves requires data. The master then generates a list of slaves from which to transfer the data to the destination slave. The master transmits the list to the resource manager. The resource manager is configured to select a source slave from the list based on available system resources. Once a source is selected by the resource manager, the master receives an instruction from the resource manager to initiate a transfer of the data from the source slave to the destination slave. The master then transmits an instruction to commence the transfer.

3. 发明公开

EP3216171A1 MULTI-PATH TIME SYNCHRONIZATION 有权
标题翻译：多路径时间同步
公开(公告)号：EP3216171A1
公开(公告)日：2017-09-13
申请号：EP14793112.5
申请日：2014-11-03
申请人： Telefonaktiebolaget LM Ericsson (publ)
发明人： RUFFINI, Stefano , PEPE, Teresa , PULERI, Marzio
IPC分类号： H04L12/26 , H04J3/06 , H04L29/06 , H04L12/24
CPC分类号： H04J3/0667 , H04J3/06 , H04L29/08135 , H04L41/12 , H04L41/142 , H04L43/00 , H04L43/0864 , H04L43/087 , H04L43/10 , H04L43/16
摘要： The method comprises analyzing (203) probing packets sent on a plurality of end-to-end paths over links of the network. At least one analysed probing packet is a timing packet comprising timing information for synchronization. The analyzing comprising analyzing probing packets sent on all of a set of probing paths. The set is selected (201) such that the analyzing provides a determination of a performance of each individual link. The method further comprises determining (204) the performance of individual links.
摘要翻译：该方法包括分析（203）探测通过网络的链路在多个端到端路径上发送的分组。至少一个分析的探测分组是包括用于同步的定时信息的定时分组。该分析包括分析在一组探测路径上发送的探测分组。该组被选择（201），使得分析提供每个单独链路的性能的确定。该方法还包括确定（204）各个链路的性能。

4. 发明公开

EP2946298A4 CONTEXT SWITCHING WITH OFFLOAD PROCESSORS 无效
标题翻译： KONTEXTUMSCHALTUNG MIT AUSGABEPROZESSOREN
公开(公告)号：EP2946298A4
公开(公告)日：2016-11-16
申请号：EP13872212
申请日：2013-06-26
申请人： XOCKETS IP LLC
发明人： DALAL PARIN , BELAIR STEPHEN
IPC分类号： H04L12/861 , G06F13/16 , G06F13/362
CPC分类号： H04L61/103 , G06F9/3877 , G06F9/461 , G06F9/4843 , G06F12/0815 , G06F12/0875 , G06F13/16 , G06F13/1652 , G06F13/285 , G06F13/362 , G06F13/4022 , G06F13/4068 , G06F15/161 , G06F15/17337 , H04L29/08135 , H04L29/08549 , H04L47/193 , H04L47/2441 , H04L47/56 , H04L47/624 , H04L47/6295 , H04L49/40 , H04L49/90 , H04L61/2592 , H04L61/6086 , Y02D10/14 , Y02D10/151
摘要： A method for context switching of multiple offload processors is disclosed. The method can include receiving network packets for processing through a memory bus connected socket, organizing the network packets into multiple sessions for processing, suspending processing of at least one session by reading a cache state of at least one of the offload processor into a context memory by operation of a scheduling circuit, with virtual memory locations and physical cache locations being aligned, and subsequently directing transfer of the cache state to at least one of the offload processors for processing by operation of the scheduling circuit.
摘要翻译：公开了一种用于多个卸载处理器的上下文切换的方法。该方法可以包括接收网络数据包以通过连接存储器总线的套接字进行处理，将网络分组组织成多个会话进行处理，通过将卸载处理器中的至少一个的高速缓存状态读取到上下文存储器中来暂停对至少一个会话的处理通过调度电路的操作，虚拟存储器位置和物理高速缓存位置被对齐，并且随后将高速缓存状态传送到至少一个卸载处理器，以通过调度电路的操作进行处理。

5. 发明公开

EP2946296A1 OFFLOAD PROCESSOR MODULES FOR CONNECTION TO SYSTEM MEMORY 无效
标题翻译：处理器模块与系统存储器结合解除
公开(公告)号：EP2946296A1
公开(公告)日：2015-11-25
申请号：EP13871993.5
申请日：2013-06-08
申请人： Xockets IP LLC
发明人： DALAL, Parin , BELAIR, Stephen
IPC分类号： G06F12/08
CPC分类号： G06F13/1652 , G06F9/3877 , G06F9/461 , G06F9/4843 , G06F12/0815 , G06F12/0875 , G06F13/16 , G06F13/285 , G06F13/362 , G06F13/4022 , G06F13/4068 , G06F15/161 , G06F15/17337 , H04L29/08135 , H04L29/08549 , H04L47/193 , H04L47/2441 , H04L47/56 , H04L47/624 , H04L47/6295 , H04L49/40 , H04L49/90 , H04L61/103 , H04L61/2592 , H04L61/6086 , Y02D10/14 , Y02D10/151
摘要： A method for context switching of multiple offload processors is disclosed. The method can include receiving network packets for processing through a memory bus connected socket, organizing the network packets into multiple sessions for processing, suspending processing of at least one session by reading a cache state of at least one of the offload processor into a context memory by operation of a scheduling circuit, with virtual memory locations and physical cache locations being aligned, and subsequently directing transfer of the cache state to at least one of the offload processors for processing by operation of the scheduling circuit.

6. 发明公开

EP2667302A1 Procédé de gestion du démarrage d'instances d'applications sur des machines virtuelles d'un réseau distribué 有权
标题翻译： Steuerungsverfahrenfürden Start mehrerer Applikationsinstanzen auf virtuellen Maschinen eines verteilten Systems
公开(公告)号：EP2667302A1
公开(公告)日：2013-11-27
申请号：EP13168389.8
申请日：2013-05-17
申请人： Schneider Electric Industries SAS
发明人： Dufrene, Pierre , Schmitt, Günther
IPC分类号： G06F9/50 , G06F9/455
CPC分类号： H04L29/08135 , G06F9/45533 , G06F9/5027 , H04L67/10
摘要： La présente invention concerne un procédé d'élection d'un emplacement virtuel pour le démarrage d'une instance d'application dans un système distribué comprenant une pluralité de machines physiques distribuées sur un réseau, chaque machine physique hébergeant au moins une machine virtuelle, et chaque machine virtuelle comprenant au moins un slot d'application, correspondant à un emplacement virtuel pour l'exécution d'une instance d'application dans ladite machine virtuelle.
Le procédé selon l'invention comprend l'élection d'un slot d'application pour le démarrage d'une instance d'application comportant les étapes suivantes :
- la détermination d'un ensemble de slots d'application libres, c'est-à-dire non occupés par une instance d'application en cours d'utilisation,
- le calcul d'un score pour chaque slot d'application libre, ledit calcul de score étant fonction d'au moins une contrainte,
- l'élection du slot d'application pour le démarrage de l'instance d'application considérée, ledit slot d'application élu étant celui présentant le meilleur score.
摘要翻译：该方法包括确定未被使用的应用程序实例占用的一组空闲应用程序段，以及根据应用程序约束计算每个空闲虚拟时隙的分数。选择具有最佳分数的虚拟插槽以启动应用实例，其中与自由虚拟时隙相关联的应用约束包括确定所请求的应用是否已经在所选择的空闲虚拟时隙中已经打开，以及授予与应用约束作为确定的函数。

7. 发明授权

EP2863308B1 Decentralized distributed computing system 有权转让
公开(公告)号：EP2863308B1
公开(公告)日：2018-08-08
申请号：EP14197207.5
申请日：2013-06-19
申请人： Hitachi, Ltd.
发明人： Hartman, David Stephen , Glass, Thomas Andrew , Sinha, Shantanu , Bernhard, Ben , Kiselev, Oleg , Mattly, James
IPC分类号： G06F9/50 , G06F9/48 , H04L12/18
CPC分类号： G06F17/30194 , G06F9/45533 , G06F9/4881 , G06F9/50 , G06F9/5005 , G06F9/5027 , G06F9/5038 , G06F9/5072 , G06F17/30091 , H04L12/1881 , H04L29/08 , H04L29/08117 , H04L29/08135 , H04L41/0803 , H04L61/2007 , H04L67/1097
摘要： A distributed system includes a plurality of nodes connected over a network. Each node has a unique identifier which identifies the node in the distributed system. Tasks are launched from any node in the distributed system and include a unique identifier. If a node receives a task including its unique identifier, the node executes the task. Each node stores each task in a log. If a node fails, after recovering from the failure, the node queries the distributed system for tasks which were not received and tasks which were not completed due to the failure. Tasks may be launched on nodes based on location data gained by using in-band access to metadata specifying a location of nodes storing data corresponding to the metadata. A task may include starting a virtual machine such that an IP address is encoded in a file configuring the virtual machine.

8. 发明公开

EP2946528A4 SCHEDULING AND TRAFFIC MANAGEMENT WITH OFFLOAD PROCESSORS 无效
标题翻译：有放电PROCESSORS规划和交通管理
公开(公告)号：EP2946528A4
公开(公告)日：2016-08-24
申请号：EP13871818
申请日：2013-06-22
申请人： XOCKETS IP LLC
发明人： DALAL PARIN , BELAIR STEPHEN
IPC分类号： H04L12/861 , G06F13/16 , H04L12/851 , H04L12/863 , H04L12/875 , H04L12/937
CPC分类号： H04L61/103 , G06F9/3877 , G06F9/461 , G06F9/4843 , G06F12/0815 , G06F12/0875 , G06F13/16 , G06F13/1652 , G06F13/285 , G06F13/362 , G06F13/4022 , G06F13/4068 , G06F15/161 , G06F15/17337 , H04L29/08135 , H04L29/08549 , H04L47/193 , H04L47/2441 , H04L47/56 , H04L47/624 , H04L47/6295 , H04L49/40 , H04L49/90 , H04L61/2592 , H04L61/6086 , Y02D10/14 , Y02D10/151
摘要： A method for context switching of multiple offload processors is disclosed. The method can include receiving network packets for processing through a memory bus connected socket, organizing the network packets into multiple sessions for processing, suspending processing of at least one session by reading a cache state of at least one of the offload processor into a context memory by operation of a scheduling circuit, with virtual memory locations and physical cache locations being aligned, and subsequently directing transfer of the cache state to at least one of the offload processors for processing by operation of the scheduling circuit.

9. 发明公开

EP2946528A2 SCHEDULING AND TRAFFIC MANAGEMENT WITH OFFLOAD PROCESSORS 无效
标题翻译： PLANUNG UND VERKEHRSVERWALTUNG MIT ENTLADUNGSPROZESSOREN
公开(公告)号：EP2946528A2
公开(公告)日：2015-11-25
申请号：EP13871818.4
申请日：2013-06-22
申请人： Xockets IP LLC
发明人： DALAL, Parin , BELAIR, Stephen
IPC分类号： H04L12/26
CPC分类号： G06F13/1652 , G06F9/3877 , G06F9/461 , G06F9/4843 , G06F12/0815 , G06F12/0875 , G06F13/16 , G06F13/285 , G06F13/362 , G06F13/4022 , G06F13/4068 , G06F15/161 , G06F15/17337 , H04L29/08135 , H04L29/08549 , H04L47/193 , H04L47/2441 , H04L47/56 , H04L47/624 , H04L47/6295 , H04L49/40 , H04L49/90 , H04L61/103 , H04L61/2592 , H04L61/6086 , Y02D10/14 , Y02D10/151
摘要： A method for context switching of multiple offload processors is disclosed. The method can include receiving network packets for processing through a memory bus connected socket, organizing the network packets into multiple sessions for processing, suspending processing of at least one session by reading a cache state of at least one of the offload processor into a context memory by operation of a scheduling circuit, with virtual memory locations and physical cache locations being aligned, and subsequently directing transfer of the cache state to at least one of the offload processors for processing by operation of the scheduling circuit.
摘要翻译：公开了一种用于多个卸载处理器的上下文切换的方法。该方法可以包括接收网络数据包以通过连接存储器总线的套接字进行处理，将网络分组组织成多个会话进行处理，通过将卸载处理器中的至少一个的高速缓存状态读取到上下文存储器中来暂停对至少一个会话的处理通过调度电路的操作，虚拟存储器位置和物理高速缓存位置被对齐，并且随后将高速缓存状态传送到至少一个卸载处理器，以通过调度电路的操作进行处理。

10. 发明公开

EP2863307A1 Decentralized distributed computing system 审中-公开
标题翻译： Dezentralisiertes verteiltes Rechnersystem
公开(公告)号：EP2863307A1
公开(公告)日：2015-04-22
申请号：EP14197198.6
申请日：2013-06-19
申请人： Hitachi Data Systems Engineering UK Limited
发明人： The designation of the inventor has not yet been filed
IPC分类号： G06F9/50 , G06F9/48 , H04L12/18
CPC分类号： G06F17/30194 , G06F9/45533 , G06F9/4881 , G06F9/50 , G06F9/5005 , G06F9/5027 , G06F9/5038 , G06F9/5072 , G06F17/30091 , H04L12/1881 , H04L29/08 , H04L29/08117 , H04L29/08135 , H04L41/0803 , H04L61/2007 , H04L67/1097
摘要： A distributed system includes a plurality of nodes connected over a network. Each node has a unique identifier which identifies the node in the distributed system. Tasks are launched from any node in the distributed system and include a unique identifier. If a node receives a task including its unique identifier, the node executes the task. Each node stores each task in a log. If a node fails, after recovering from the failure, the node queries the distributed system for tasks which were not received and tasks which were not completed due to the failure. Tasks may be launched on nodes based on location data gained by using in-band access to metadata specifying a location of nodes storing data corresponding to the metadata. A task may include starting a virtual machine such that an IP address is encoded in a file configuring the virtual machine.
摘要翻译：分布式系统包括通过网络连接的多个节点。每个节点都有唯一的标识符，用于标识分布式系统中的节点。任务从分布式系统中的任何节点启动，并包括一个唯一的标识符。如果节点接收到包含其唯一标识符的任务，则节点执行该任务。每个节点将每个任务存储在日志中。如果节点出现故障，则在从故障恢复后，节点向分布式系统查询未接收的任务，以及由于故障而未完成的任务。可以基于通过使用对指定与元数据相对应的数据的节点的位置的元数据的元数据访问获得的位置数据在节点上启动任务。任务可以包括启动虚拟机，使得IP地址被编码在配置虚拟机的文件中。

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