some observations -
1. "spa_usage1 contains app 54 million rows" vs. "The size of Spool 2 is estimated with low confidence to be 567,176,439 rows"
which is a 10x difference. Stats not up to date? or wrong expectations?
2. PPI "PARTITION BY RANGE_N(CAST((USG_DT_SK ) AS INTEGER) BETWEEN 17898 AND 18170 EACH 1, NO RANGE)" does not support at least this query well...
3. PI (S_SK,C_SK) indicate multiple rows for S_SK. Do you have highly / extreme skewed values?
↧