Rebuilding indexes with monotonically increasing sequences

Question

Hi
I have been reading this paper
http://richardfoote.files.wordpress.com/2007/12/index-internals-rebuilding-the-truth.pdf
*On page 91 we  have*
CREATE TABLE test_empty_block (id NUMBER, name VARCHAR2(30));
Table created.
INSERT INTO test_empty_block SELECT rownum, 'BOWIE' FROM dual
CONNECT BY level <= 10000;
10000 rows created.
COMMIT;
CREATE INDEX test_empty_block_idx ON test_empty_block(id);
Index created.
On page 92 we have
SQL> DELETE test_empty_block WHERE id between 1 and 9990;
9990 rows deleted.
SQL> COMMIT;
Commit complete.
SQL> ANALYZE INDEX test_empty_block_idx VALIDATE STRUCTURE;
Index analyzed.
SQL> SELECT lf_blks, del_lf_rows FROM index_stats;
LF_BLKS DEL_LF_ROWS
21 9990
On page 93
SQL> INSERT INTO test_empty_block SELECT rownum+20000, 'ZIGGY' FROM dual CONNECT BY level <= 10000;
10000 rows created.
SQL> COMMIT;Commit complete.
SQL> ANALYZE INDEX test_empty_block_idx VALIDATE STRUCTURE;
Index analyzed.
SQL> SELECT lf_blks, del_lf_rows FROM index_stats;
LF_BLKS DEL_LF_ROWS
21 0
when the DEL_LF_ROWS is 0
How come Richard goes on to state that on page 96.
Although deleted index space is generally reusable, there can be wasted space
Monotonically increasing index values and deletions
Edited by: 899881 on Jan 3, 2012 9:05 PM
Edited by: 899881 on Jan 3, 2012 9:06 PM

Answer 1

899881 wrote:
How come Richard goes on to state that on page 96.
Although deleted index space is generally reusable, there can be wasted space
Monotonically increasing index values and deletions
I believe this is addressed in page 103 of the same document you posted a link to.