Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Accessing Rows Based on Hidden Field Values

IP.com Disclosure Number: IPCOM000110430D
Original Publication Date: 1992-Nov-01
Included in the Prior Art Database: 2005-Mar-25
Document File: 1 page(s) / 54K

Publishing Venue

IBM

Related People

Chen, Y: AUTHOR [+3]

Abstract

To reorganize or add nodes to a partitioned database, an efficient method of accessing the rows to be moved is needed. A hidden field containing the bucket number and a hidden field giving a row's address or ID (RID) can be exploited for this purpose, as outlined in this invention. 1. Create and maintain an index on the bucket number (BNUM) hidden field. The bucket number refers to the result of hashing a partitioning key value in the rows of the table, determining which partition each row of the table will be placed in. There can be many more buckets than nodes, each node receiving the rows of several buckets. 2. The index maps BNUM to RID for each row in the partitioned table. 3.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 77% of the total text.

Accessing Rows Based on Hidden Field Values

      To reorganize or add nodes to a partitioned database, an
efficient method of accessing the rows to be moved is needed.  A
hidden field containing the bucket number and a hidden field giving a
row's address or ID (RID) can be exploited for this purpose, as
outlined in this invention.
1.  Create and maintain an index on the bucket number (BNUM) hidden
field.  The bucket number refers to the result of hashing a
partitioning key value in the rows of the table, determining which
partition each row of the table will be placed in.  There can be many
more buckets than nodes, each node receiving the rows of several
buckets.
2.  The index maps BNUM to RID for each row in the partitioned table.
3.  To access all the rows belonging to a given bucket or a given set
of buckets, search the index using the given bucket number(s) as
search key(s).  The database management system's index manager is
used for these scans.
4.  The RIDs that are returned from the scan should be sorted, for
the most efficient retrieval.  This can be done by using the database
management system's sort services in the normal manner, with the RID
column being the sort key.
5.  The sorted list of RIDs is then consulted, and each row pointed
to by each RID is retrieved, using the database management system's
record retrieval service.

      Benefits of using this technique:
1.  Good performance: an index lookup prevents the need to access
rows that a...