thanks
Printable View
thanks
what is the differentiate between use:
analyze table xxx subpartition yyy compute statistics
and
dbms_stats.gather_table_stats('zzz','table_name','partition_name')
Why the second option runs quickly and the first option runs very slowly.
That's why you need to see why dbms_stats is choosen against analyze table command.
The procedural nature, and the fact that dbms_stats directly collects stats on part. tables makes it work efficiently. analyze does the global stas which can also lead to inaccuracies.
So dump the analyze and keep using dbms_stats, in 9i there are more added advantages of dbms_stats, i suggest read some documentation to see the high difference these two tools may have.
but if I have analyzed partitionated tables using command analyze and now I use this plsql, I haven´t to delete previous statistics.
Is that Ok??