发明授权
US08478775B2 Efficient large-scale filtering and/or sorting for querying of column based data encoded structures
有权
用于查询基于列的数据编码结构的高效大规模过滤和/或排序
- 专利标题: Efficient large-scale filtering and/or sorting for querying of column based data encoded structures
- 专利标题(中): 用于查询基于列的数据编码结构的高效大规模过滤和/或排序
-
申请号: US12363637申请日: 2009-01-30
-
公开(公告)号: US08478775B2公开(公告)日: 2013-07-02
- 发明人: Amir Netz , Cristian Petculescu , Adrian Ilcu Predescu , Marius Dumitru
- 申请人: Amir Netz , Cristian Petculescu , Adrian Ilcu Predescu , Marius Dumitru
- 申请人地址: US WA Redmond
- 专利权人: Microsoft Corporation
- 当前专利权人: Microsoft Corporation
- 当前专利权人地址: US WA Redmond
- 代理机构: Workman Nydegger
- 主分类号: G06F17/00
- IPC分类号: G06F17/00 ; G06F7/00
摘要:
The subject disclosure relates to querying of column based data encoded structures enabling efficient query processing over large scale data storage, and more specifically with respect to complex queries implicating filter and/or sort operations for data over a defined window. In this regard, in various embodiments, a method is provided that avoids scenarios involving expensive sorting of a high percentage of, or all, rows, either by not sorting any rows at all, or by sorting only a very small number of rows consistent with or smaller than a number of rows associated with the size of the requested window over the data. In one embodiment, this is achieved by splitting an external query request into two different internal sub-requests, a first one that computes statistics about distribution of rows for any specified WHERE clauses and ORDER BY columns, and a second one that selects only the rows that match the window based on the statistics.
公开/授权文献
信息查询