Re: Storage Structures clarification

From: Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk>
Date: Tue, 23 Mar 1999 19:58:33 -0000
Message-ID: <922219182.8935.2.nnrp-10.9e984b29@news.demon.co.uk>

The ambiguity comes from INDEX CLUSTERS. A cluster is also a single segment, but a cluster can contain main tables. The table directory in a cluster block lists the tables that appear in that block.

In fact, the row id for a row in a clustered table is (invisibly) longer than the rowid of a non-clustered table, because the table's index-id is appended to the rest of the rowid.

Jonathan Lewis
Yet another Oracle-related web site: www.jlcomp.demon.co.uk

Joe gulla wrote in message <36F685A3.FE233C77_at_erosl.com>...
> I'm slightly confused . If , out of manual ,"each table's data is
>store in it's own segment " sounds singular to me. Segment made up of
>extents, extents made up
>of blocks.
> The block header contains "table directory" which contain's
>information about
>tables having rows in the block. This sounds like more than one table
>can
>have rows in a block.
> Is this correct ?
> Joe Gulla
>
>
Received on Tue Mar 23 1999 - 13:58:33 CST