知搜-全球专利检索

  1. 登录
  2. 注册

搜索结果

466 条记录 (耗时 0.028 秒)

    METHOD TO STORE DATA ASSOCIATED TO MEMBERS OF ENTITIES OF A DATABASE
    1.
    发明公开
    METHOD TO STORE DATA ASSOCIATED TO MEMBERS OF ENTITIES OF A DATABASE 审中-公开

    公开(公告)号：US20240354294A1

    公开(公告)日：2024-10-24

    申请号：US18580666

    申请日：2021-07-29

    申请人： Franco ROSI ,  Natzka SA

    发明人： Franco ROSI

    IPC分类号： G06F16/22

    CPC分类号： G06F16/2282 ,  G06F16/2237

    摘要： A method for organising data into data structures and accessing data in those structures is described. Starting with a tabular structure, each row is coded as a first and second macro number, which are stored in a configuration structure. A pre-mapping is captured from the data structure, wherein the second macro numbers are stored without repetition in a pre-mapping structure with segments i, with a number K of second numbers. A pointing function to the second macro numbers from the first macro numbers is also implemented via bitmaps with K bits, wherein a bitmap i is allocated for a first macro number only if the first macro number in the configuration structure is in a row that also contains the second macro number, and the second macro number is in segment i in the pre-mapping structure. A method for accessing data stored in the aforementioned structures is also described.

    Methods for verifying vector database query results and devices thereof
    3.
    发明授权
    Methods for verifying vector database query results and devices thereof 有权

    公开(公告)号：US12117995B1

    公开(公告)日：2024-10-15

    申请号：US18620214

    申请日：2024-03-28

    申请人： SPACE AND TIME LABS, INC.

    发明人： Jay Thomas White ,  Scott Edward Daly Dykstra

    IPC分类号： G06F16/23 ,  G06F16/22 ,  G06F16/2455

    CPC分类号： G06F16/2379 ,  G06F16/2237 ,  G06F16/2455

    摘要： Methods, non-transitory computer-readable media, and query verification apparatuses are disclosed that receive a result vector, an index to the result vector, a range proof, and an inner product proof from a prover device after providing a query vector to the prover device. The prover device hosts a vector database with rows and columns. Second and third commitments to a distance vector and a random linear combination vector, respectively, are generated based on a first commitment to the rows. The range proof is checked using the second commitment and the inner product proof is checked using the third commitment and a generator used to generate the first commitment and identified based on the index. The result vector is returned after confirming that it is one of the rows and each of the rows is as far from the query vector as the result vector based on the check of the proofs.

    Table scan predicate with integrated semi-join filter
    6.
    发明授权
    Table scan predicate with integrated semi-join filter 有权

    公开(公告)号：US12099490B2

    公开(公告)日：2024-09-24

    申请号：US17751296

    申请日：2022-05-23

    申请人： SAP SE

    发明人： Manuel Mayr ,  Wolfgang Stephan ,  Till Merker

    IPC分类号： G06F16/22 ,  G06F16/242 ,  G06F16/2455

    CPC分类号： G06F16/2282 ,  G06F16/221 ,  G06F16/2237 ,  G06F16/2428 ,  G06F16/2456

    摘要： A table scan predicate with integrated semi-join filter is provided. A method includes receiving a query including: a request to join first data from a first table and second data from a second table, a first predicate for use in a table scan of the second table, and a second predicate including an expression associated with the first data from the first table and a reference to a column associated with the second data from the second table. The method may include transforming the second predicate into a dynamic predicate for execution of the query. the method may include applying the dynamic predicate to at least the first data. The method may include executing the query by at least scanning the second table based on the first predicate and filtered first data from the application of the dynamic predicate. Related systems and articles of manufacture are provided.

    Techniques for executing join operations using bitmap indices
    9.
    发明授权
    Techniques for executing join operations using bitmap indices 有权

    公开(公告)号：US12086136B2

    公开(公告)日：2024-09-10

    申请号：US16782025

    申请日：2020-02-04

    申请人： DISNEY ENTERPRISES, INC.

    发明人： Guy Molinari

    IPC分类号： G06F16/2453 ,  G06F16/22 ,  G06F16/2455

    CPC分类号： G06F16/24537 ,  G06F16/2237 ,  G06F16/24532 ,  G06F16/24544 ,  G06F16/2456

    摘要： A database engine uses bitmap slice indices (BSIs) to combine result bitmaps derived from parent and child tables related via one or more foreign key relationships. To execute a query that involves a parent table and a child table, the database engine parses the query into a parent predicate corresponding to the parent table and a child predicate corresponding to the child table. The database engine executes the parent predicate using a parent bitmap index to generate a parent result bitmap and executes the child predicate using a child bitmap index to generate a child result bitmap. The database engine transforms the child result bitmap, based on the BSI, to generate a transformed child result bitmap having columns that correspond to the primary key of the parent table. The database engine can then combine the parent result bitmap directly with the transformed child result bitmap to generate query results.

    Compression for sparse data structures utilizing mode search approximation
    10.
    发明授权
    Compression for sparse data structures utilizing mode search approximation 有权

    公开(公告)号：US12086120B2

    公开(公告)日：2024-09-10

    申请号：US18066436

    申请日：2022-12-15

    申请人： Intel Corporation

    发明人： Prasoonkumar Surti ,  Abhishek R. Appu ,  Karol Szerszen ,  Eric Liskay ,  Karthik Vaidyanathan

    IPC分类号： G06F16/22 ,  G06N20/00 ,  G06T1/20

    CPC分类号： G06F16/2237 ,  G06N20/00 ,  G06T1/20

    摘要： Embodiments are generally directed to compression for compression for sparse data structures utilizing mode search approximation. An embodiment of an apparatus includes one or more processors including a graphics processor to process data; and a memory for storage of data, including compressed data. The one or more processors are to provide for compression of a data structure, including identification of a mode in the data structure, the data structure including a plurality of values and the mode being a most repeated value in a data structure, wherein identification of the mode includes application of a mode approximation operation, and encoding of an output vector to include the identified mode, a significance map to indicate locations at which the mode is present in the data structure, and remaining uncompressed data from the data structure.