|
|
Archives of the TeradataForum
Message Posted: Wed, 20 Jun 2012 @ 14:29:37 GMT
Subj: | | Re: Collect stats on VT in TD13.10 |
|
From: | | Md, Asifuddin |
Q)
We know in TD 12 and below versions there is no option to collect stats on Volatile table.
But I heard we can collect stats on VT in TD 13 and above.
A)
Sugesh:
Syntaxes can be checked with below once you are on Database:
help 'sql collect statistics';
And currently there is no option directly provided on VT as I see on 13.10.
help 'sql collect statistics';
*** Query completed. 93 rows found. One column returned.
*** Total elapsed time was 1 second.
On-Line Help
---------------------------------------------------------------------------
/* OPTIMIZER FORM */
{ COLLECT STATISTICS }
{ COLLECT STATS } [ USING SAMPLE ] [ ON ]
{ COLLECT STAT }
{ [ TEMPORARY ] [ database_name. ] table_name }
{ [ database_name. ] join_index_name }
{ [ database_name. ] hash_index_name }
[ COLUMN { column_name } ]
{ PARTITION }
[ COLUMN ({ column_name } [ ,PARTITION ] [ ..64,column_name ] ) ];
({ PARTITION } [ ..64,column_name ] )
[ INDEX ( column_name [..64,column_name ] ) ]
[ INDEX index_name ]
[ FROM [TEMPORARY] tablename ]
[{ index_description[ ...,index_description ] }
{ column_description[ ...,column_description ] }];
OR
/* QCD FORM */
{COLLECT STATISTICS}
{COLLECT STATS } { FOR SAMPLE [PERCENT] INTO }
{COLLECT STAT }
[SET QUERY (query_id ) ][SAMPLEID (statistics_id)] [UPDATE MODIFIED][ON]
[ database_name. ] table_name
{ COLUMN { column_name } }
{ PARTITION }
{ COLUMN ( { column_name } [,PARTITION] [ ..64,column_name ] ) };
( { PARTITION } [ ..64,column_name ] )
{ INDEX ( column_name [..64,column_name ] ) }
{ INDEX index_name }
where
Percentage
The sample data size as a percentage of the total set of rows on
a given AMP. It is a float value and the value must be less than 10
QCDName
Name of the database where QCF tables reside
OR
/* "CREATE INDEX" - style syntax */
{COLLECT STATISTICS}
{COLLECT STATS } [ USING SAMPLE ]
{COLLECT STAT }
{ index_description[ ...,index_description ] } ON
{ column_description[ ...,column_description ] }
{ [ TEMPORARY ][ database_name. ] table_name }
{ [ database_name. ] join_index_name };
{ [ database_name. ] hash_index_name }
[ FROM [TEMPORARY] tablename ]
[{ index_description[ ...,index_description ] }
{ column_description[ ...,column_description ] }];
where
index_description:
[ UNIQUE ] INDEX [ index_name ] [ ALL ]
( column_name1 [ ..64,column_name1 ] )
[ ORDER BY [ VALUES | HASH ](column_name_2) ]
column_description:
COLUMN { column_name }
{ PARTITION }
{({ column_name } [ ,PARTITION ] [ ..64,column_name ])
{({ PARTITION } [ ..64,column_name ])
ALL is used for secondary indexes created on join indexes only.
Thanks
Asif - TD
| |