公开(公告)号：US5978792A

公开(公告)日：1999-11-02

申请号：US6745

申请日：1998-01-13

Applicant: Gautam Bhargava ,  Paramesh Sampatrai Desai ,  Piyush Goel ,  Peter Hoa ,  Fen-Ling Lin ,  Balakrishna Raghavendra Iyer ,  Jerry Mukai ,  William Samuel Perlman ,  Hong Sang Tie

Inventor： Gautam Bhargava ,  Paramesh Sampatrai Desai ,  Piyush Goel ,  Peter Hoa ,  Fen-Ling Lin ,  Balakrishna Raghavendra Iyer ,  Jerry Mukai ,  William Samuel Perlman ,  Hong Sang Tie

IPC: G06F17/30

CPC classification number: G06F17/30327 ,  Y10S707/99931 ,  Y10S707/99932 ,  Y10S707/99933 ,  Y10S707/99934 ,  Y10S707/99936 ,  Y10S707/99937

Abstract: A method and apparatus for generating static, dynamic and hybrid sparse indices for use with workfiles used by SQL queries in a relational database management system.

Abstract translation: 一种用于生成关系数据库管理系统中SQL查询使用的工作文件的静态，动态和混合稀疏索引的方法和装置。

公开(公告)号：US5758145A

公开(公告)日：1998-05-26

申请号：US393803

申请日：1995-02-24

Applicant: Gautam Bhargava ,  Paramesh Sampatrai Desai ,  Piyush Goel ,  Peter Hoa ,  Fen-Ling Lin ,  Balakrishna Raghavendra Iyer ,  Jerry Mukai ,  William Samuel Perlman ,  Hong Sang Tie

Inventor： Gautam Bhargava ,  Paramesh Sampatrai Desai ,  Piyush Goel ,  Peter Hoa ,  Fen-Ling Lin ,  Balakrishna Raghavendra Iyer ,  Jerry Mukai ,  William Samuel Perlman ,  Hong Sang Tie

IPC: G06F17/30

CPC classification number: G06F17/30327 ,  Y10S707/99931 ,  Y10S707/99932 ,  Y10S707/99933 ,  Y10S707/99934 ,  Y10S707/99936 ,  Y10S707/99937

Abstract: A method, apparatus and article for manufacture for generating static, dynamic and hybrid sparse indices for use with workfiles used by SQL queries in a relational database management system. A workfile and a sparse index structure are temporarily created in the computer during execution of the query by the computer. The workfile stores intermediate relations resulting from execution of a portion of the SQL query, wherein the intermediate relations comprise sorted rows for an inner table referenced in the SQL query. The sparse index structure contains one or more entries indicating at least an approximate location for at least some of the rows in the workfile. As one or more rows from an outer table referenced in the SQL query are retrieved, the entries of the sparse index structure are searched for a closest matching entry for each retrieved row. The sorted workfile is then scanned for a row matching each retrieved row using the closest matching entry from the sparse index structure as a starting position. The sparse index structure is then updated with an entry corresponding to the row from the sorted workfile matching the retrieved row from the outer table.

Abstract translation: 一种用于生成静态，动态和混合稀疏索引的方法，装置和制品，用于在关系数据库管理系统中用于SQL查询的工作文件。 在计算机执行查询期间，在计算机中临时创建工作文件和稀疏索引结构。 工作文件存储由执行SQL查询的一部分而产生的中间关系，其中中间关系包括在SQL查询中引用的内部表的排序行。 稀疏索引结构包含一个或多个条目，至少指示工作文件中至少一些行的近似位置。 当从SQL查询中引用的外部表中的一行或多行被检索到时，搜索稀疏索引结构的条目以获取每个检索行的最接近的匹配条目。 然后，使用距离稀疏索引结构最接近的匹配条目作为起始位置，扫描排序的工作文件，使其与每个检索到的行相匹配的行。 然后使用与来自外部表的检索到的行匹配的排序工作文件中的行对应的条目来更新稀疏索引结构。