利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

173 条记录 (耗时 0.019 秒)

    SYSTEM AND METHOD FOR MANAGING DISTRIBUTION OF COMPUTATIONS IN MULTI-ANTENNA AND MULTI-TRANSMITTER ENVIRONMENTS
    4.
    发明公开
    SYSTEM AND METHOD FOR MANAGING DISTRIBUTION OF COMPUTATIONS IN MULTI-ANTENNA AND MULTI-TRANSMITTER ENVIRONMENTS 审中-公开

    公开(公告)号：EP3496366A1

    公开(公告)日：2019-06-12

    申请号：EP17290156.3

    申请日：2017-12-05

    申请人： Institut Eurecom G.I.E.

    发明人： Elia, Petros ,  Lampiris, Eleftherios ,  Parrinello, Emanuele

    IPC分类号： H04L29/08 ,  G06F7/24 ,  G06F9/50 ,  H04L1/00

    摘要： In the field of distributed computing, the communication that needs to be performed between the computing nodes or helper nodes, poses a major performance bottleneck. In order to reduce the time associated with this so-called shuffling phase of inter-node communication, it was recently proposed that the computational redundancy across the nodes be exploited to allow - via a specific type of coded communications - that each transmission becomes useful to more than one node at a time, thus significantly reducing the overall communication time. But the authors in this patent application here, have identified as a main bottleneck of such coded' current approaches, the fact that the input dataset to be processed for a given job, has to be split into an exponentially large number of smaller subsets, which is often impossible, thus significantly constraining the speedup gains attributed to coded communication. The inventors here have invented a new method of assigning data to the computing nodes, which dramatically alleviates the aforementioned bottleneck, thus resulting in dramatic reduction to the overall execution time of the distributed computing job.

    PARALLEL PRIORITY QUEUE UTILIZING PARALLEL HEAP ON MANY-CORE PROCESSORS FOR ACCELERATING PRIORITY-QUEUE-BASED APPLICATIONS
    5.
    发明公开
    PARALLEL PRIORITY QUEUE UTILIZING PARALLEL HEAP ON MANY-CORE PROCESSORS FOR ACCELERATING PRIORITY-QUEUE-BASED APPLICATIONS 审中-公开
    并行优先级队列中通过平行堆上基于优先级队列中加速应用的多核处理器

    公开(公告)号：EP2936308A4

    公开(公告)日：2016-07-20

    申请号：EP13864943

    申请日：2013-12-19

    申请人： UNIV GEORGIA STATE RES FOUND

    发明人： PRASAD SUSHIL K ,  HE XI ,  AGARWAL DINESH

    IPC分类号： G06F9/46 ,  G06F7/00 ,  G06F7/24 ,  G06F9/38 ,  G06F9/48 ,  G06F9/52 ,  G06F9/54 ,  G06T1/20

    CPC分类号： G06F9/522 ,  G06F7/24 ,  G06F9/4881 ,  G06F9/545 ,  G06F9/546 ,  G06F2209/548 ,  G06T1/20

    摘要： Disclosed are various embodiments for a parallel priority queue implemented on one or more many-core processors and/or multi-core processors such as those in general-purpose graphics processing units (GPGPUs). According to various embodiments, a priority may be determined according to a timestamp of an item, such as an event or an entry, in a priority queue. A priority queue interface may comprise functions to insert and remove entries from the priority queue. Priority order of the entries may be maintained as the entries are inserted and removed from the queue.

    摘要翻译： 公开的是用于实现在一个或多个多核心处理器和/或多核处理器并行优先级队列各种实施例：诸如那些在通用的图形处理单元（GPGPU）。 。根据各种实施例，优先级可以是确定性的开采gemäß到项目的时间戳，作为寻求事件或在一个优先级队列中的条目。 优先级队列接口可以包括函数来插入并从优先级队列中删除条目。 作为条目被插入并从队列中删除的条目的优先顺序可以保持。

    SORTING
    7.
    发明公开
    SORTING 有权
    分拣

    公开(公告)号：EP2676189A2

    公开(公告)日：2013-12-25

    申请号：EP12705604.2

    申请日：2012-02-16

    申请人： Ab Initio Technology LLC

    发明人： STANFILL, Craig, W. ,  FEYNMAN, Carl, Richard

    IPC分类号： G06F7/24

    CPC分类号： G06F7/24 ,  G06F7/22 ,  G06F9/544 ,  G06F9/546 ,  G06F17/30312 ,  G06F2207/226 ,  G06F2209/548

    摘要： Systems and techniques are disclosed that include in one aspect a computer implemented method storing a received stream of data elements in a buffer, applying a boundary condition to the data elements stored in the buffer after receiving each individual data element of the stream of data elements, and producing one or more data elements from the buffer based on the boundary condition as an output stream of data elements sorted according to a predetermined order.

    摘要翻译： 公开了系统和技术，其在一个方面包括计算机实现的方法，该方法将接收到的数据元素流存储在缓冲器中，在接收到数据元素流的每个单独数据元素之后，将边界条件应用于存储在缓冲器中的数据元素， 以及基于所述边界条件从所述缓冲器产生一个或多个数据元素作为根据预定顺序排序的数据元素的输出流。

    A method of querying a data structure
    8.
    发明公开
    A method of querying a data structure 审中-公开
    Verfahren zur Abfrage einer Datenstruktur

    公开(公告)号：EP2618275A1

    公开(公告)日：2013-07-24

    申请号：EP12194308.8

    申请日：2012-11-26

    申请人： Data-Re Limited

    发明人： Longshaw, Thomas, Benjamin

    IPC分类号： G06F17/30 ,  G06F7/24

    CPC分类号： G06F17/30321 ,  G06F7/24 ,  G06F17/30595

    摘要： There is provided a method of querying a data structure (100), the data structure comprising a plurality of records (T1 - T7) and a plurality of data items (R, G, B, Y, O), each record comprising an index identifying the record and a collection of paths (P1 - P5) pointing to a selection of the data items. Each path of a record corresponds to a respective data variable, the value of the data variable being the data item pointed to by the path. The method comprises receiving (200) a query to sort a sub-set of the records according to a data variable; creating (210) a first array; defining (220) start points in the first array for respective data items of the first data variable according to the order in which the data items are to be sorted; for each record of the sub-set, storing (230) the index of the record after the start point corresponding to the data item of the first data variable of the record; determining (240) an end point for each start point; determining (250) which locations in each sequence of locations within the first array where indexes to the records have been stored based on the start points and the end points; and returning (250) the determined locations to designate where the record indexes corresponding to the respective data items may be found.

    摘要翻译： 提供了一种查询数据结构（100）的方法，所述数据结构包括多个记录（T1-T7）和多个数据项（R，G，B，Y，O），每个记录包括索引 识别记录和指向数据项的选择的路径（P1-P5）的集合。 记录的每个路径对应于相应的数据变量，数据变量的值是路径指向的数据项。 该方法包括：根据数据变量接收（200）查询以对记录的子集进行排序; 创建（210）第一阵列; 根据要对数据项进行排序的顺序，为第一数据变量的各个数据项定义（220）起始点; 对于子集的每个记录，存储（230）与记录的第一数据变量的数据项对应的起始点之后的记录的索引; 确定（240）每个起点的终点; 确定（250）基于起始点和终点，在第一阵列内的每个记录序列中哪些位置已经存储了对记录的索引; 并返回（250）确定的位置，以指定可以找到与各个数据项对应的记录索引的位置。

    Multiple key self-sorting table
    10.
    发明公开
    Multiple key self-sorting table 审中-公开
    自排序表有多个键的

    公开(公告)号：EP1482403A3

    公开(公告)日：2007-03-21

    申请号：EP04252867.9

    申请日：2004-05-17

    申请人： Ericsson AB

    发明人： Hook, Joseph A. ,  Horner, Jeremy William

    IPC分类号： G06F7/24

    CPC分类号： G06F7/24

    摘要： A multiple key self-sorting table. The table includes a plurality of intelligent self-sorting modules. Each module of the plurality of modules having an entry with at least one value. Each module making a decision to perform one of a plurality of acts, and preferably four acts, when there is an addition or deletion of an entry to the table; it holds its current value, stores a new entry, takes an entry from its immediately adjacent module with a higher value or takes an entry from its immediately adjacent module with a lower value; and performing the decision. A method for sorting a table. The method includes the steps of making a decision by each module, having an entry with at least one value, of a plurality of modules intelligent self-sorting modules to perform one of a plurality of acts, and preferably four acts, when there is an addition or deletion of an entry to the table; holding its current value, storing a new entry, taking an entry from its immediately adjacent module with a higher value or taking an entry from its immediately adjacent module with a lower value. There is the step of performing the decision.