利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

5 条记录 (耗时 0.031 秒)

    Sparse Linear Algebra in Column-Oriented In-Memory Database
    3.
    发明申请
    Sparse Linear Algebra in Column-Oriented In-Memory Database 审中-公开
    面向列的内存数据库中的稀疏线性代数

    公开(公告)号：US20150379054A1

    公开(公告)日：2015-12-31

    申请号：US14314750

    申请日：2014-06-25

    申请人： David Kernert ,  Frank Koehler ,  Wolfgang Lehner

    发明人： David Kernert ,  Frank Koehler ,  Wolfgang Lehner

    IPC分类号： G06F17/30

    CPC分类号： G06F17/16 ,  G06F17/30153

    摘要： Embodiments relate to storing sparse matrices in an in-memory column-oriented database system. Specifically, recent hardware shifts of primary storage from disc into memory, allow execution of linear algebra queries directly in the database engine. Dynamic matrix manipulation operations (like online insertion or deletion of elements) are not covered by most linear algebra frameworks. Therefore a hybrid architecture comprises a read-optimized main structure, and a write-optimized delta structure. The resulting system layout derived from the Compressed Sparse Row (CSR) representation, integrates well with a columnar database design. Moreover, the resulting architecture is amenable to a wide range of non-numerical use cases when dictionary encoding is used. Performance in specific examples is evaluated for dynamic sparse matrix workloads, by applying work flows of nuclear science and network graphs. Embodiments allow performing linear algebra operations on large, sparse matrices commonly associated with scientific computations and analytical business applications.

    摘要翻译： 实施例涉及将存储稀疏矩阵存储在面向内存的列的数据库系统中。 具体来说，最近硬盘将主存储从盘转移到存储器中，允许直接在数据库引擎中执行线性代数查询。 动态矩阵处理操作（如在线插入或删除元素）不被大多数线性代数框架所涵盖。 因此，混合架构包括读取优化的主要结构和写入优化的三角形结构。 从压缩稀疏行（CSR）表示派生的生成的系统布局与一个柱状数据库设计相结合。 而且，当使用字典编码时，得到的结构适用于广泛的非数值使用情况。 通过应用核科学和网络图的工作流量，对具体示例中的性能进行动态稀疏矩阵工作负载的评估。 实施例允许对通常与科学计算和分析业务应用相关联的大的稀疏矩阵执行线性代数运算。

    Combination filter for filtering fluids
    4.
    发明授权
    Combination filter for filtering fluids 失效
    用于过滤液体的组合过滤器

    公开(公告)号：US06986804B2

    公开(公告)日：2006-01-17

    申请号：US10472785

    申请日：2002-04-05

    申请人： Klaus Dominiak ,  Frank Koehler ,  Nicolai Bochynek ,  Marcus Lotgerink-Bruinenberg

    发明人： Klaus Dominiak ,  Frank Koehler ,  Nicolai Bochynek ,  Marcus Lotgerink-Bruinenberg

    IPC分类号： B01D29/07 ,  B01D46/12 ,  B01D53/04

    CPC分类号： B01D46/521 ,  B01D46/0036 ,  B01D46/10 ,  B01D46/2411 ,  B01D53/02 ,  B01D53/0446 ,  B01D2253/102 ,  B01D2253/342 ,  B01D2275/10

    摘要： The present invention provides a combination filter for filtering fluids comprising a flow channel particulate filtration media having a first face and a second face and a gas adsorbing filtration media. The flow channel particulate filtration media comprises a plurality of flow channels directed in flow direction and defined by inner surfaces. The flow channels having inlet openings through the first face and outlet openings through the second face of the flow channel particulate filtration media. The inner surfaces of said flow channels at least in part are provided with structures protruding therefrom and forming or extending into the flow channels or an electrical charge or a combination of both. The said gas adsorbing filtration media comprises a pad having a first face and a second face and width and length dimensions orthogonal with respect to each other and each individually to the flow direction and having a thickness dimension in flow direction. The pad comprising a layer extending substantially perpendicular to the flow direction across the width and length dimensions of the pad.

    摘要翻译： 本发明提供一种用于过滤流体的组合过滤器，其包括具有第一面和第二面的流动通道微粒过滤介质和气体吸附过滤介质。 流动通道颗粒过滤介质包括沿流动方向引导并由内表面限定的多个流动通道。 流动通道具有穿过第一面的入口和穿过流动通道微粒过滤介质的第二面的出口。 所述流动通道的内表面至少部分地设置有从其突出的结构，并形成或延伸到流动通道或电荷或两者的组合中。 所述气体吸附过滤介质包括具有第一面和第二面的垫，宽度和长度尺寸相对于彼此正交，并且各自独立于流动方向并具有沿流动方向的厚度尺寸。 垫包括穿过垫的宽度和长度尺寸基本垂直于流动方向延伸的层。

    Sparse linear algebra in column-oriented in-memory database
    5.
    发明授权
    Sparse linear algebra in column-oriented in-memory database 有权

    公开(公告)号：US10067909B2

    公开(公告)日：2018-09-04

    申请号：US14314750

    申请日：2014-06-25

    申请人： David Kernert ,  Frank Koehler ,  Wolfgang Lehner

    发明人： David Kernert ,  Frank Koehler ,  Wolfgang Lehner

    IPC分类号： G06F17/16 ,  G06F17/30

    摘要： Embodiments relate to storing sparse matrices in an in-memory column-oriented database system. Specifically, recent hardware shifts of primary storage from disc into memory, allow execution of linear algebra queries directly in the database engine. Dynamic matrix manipulation operations (like online insertion or deletion of elements) are not covered by most linear algebra frameworks. Therefore a hybrid architecture comprises a read-optimized main structure, and a write-optimized delta structure. The resulting system layout derived from the Compressed Sparse Row (CSR) representation, integrates well with a columnar database design. Moreover, the resulting architecture is amenable to a wide range of non-numerical use cases when dictionary encoding is used. Performance in specific examples is evaluated for dynamic sparse matrix workloads, by applying work flows of nuclear science and network graphs. Embodiments allow performing linear algebra operations on large, sparse matrices commonly associated with scientific computations and analytical business applications.