Degree in gather procedures in package DBMS_STATS

Hi all,

My DB is 11.1 and my collection procedure looks like:

BEGIN
  DBMS_STATS.GATHER_TABLE_STATS(ownname          => user,
                                tabname          => 'S_BK',
                                method_opt       => 'for all columns size auto',
                                degree           => 4,
                                estimate_percent =>10,
                                granularity      => 'ALL',
                                cascade          => TRUE);
END;
/

It costs 5 mins to complete. I aslo test with setting degree to NULL and degree to 8, but the total execution times have no differences.

What's the use of DGREE parameter here?

I know what the degree parameter works for according to Oracle document. But after testing, I found that bigger degree didn't make the process faster.

Please note I did the test at the idle time of the database when my DB has little work to do and has sufficient processing power.