Archives of the TeradataForum
Message Posted: Thu, 24 Mar 2011 @ 12:45:10 GMT
I need some help regarding improving performance on big table we have EVENT_GL_AMOUNT table size around 1.101 tB and with nearly rows 1 The table is partitioned on EVENT_AMOUNT_TYPE_CD however the joins and other performance is poor , even normal select and counts affecting performance please guide what else we can do to improve performance from this table.
Table definition is as follows
CREATE SET TABLE DW.EVENT_GL_AMOUNT ,NO FALLBACK , NO BEFORE JOURNAL, NO AFTER JOURNAL, CHECKSUM = DEFAULT ( Event_Id DECIMAL(18,0), Event_Amount_Type_Cd SMALLINT, Event_Amount_Val DECIMAL(18,2) COMPRESS , GL_Account_Id INTEGER COMPRESS , Unit_Of_Measure_Cd SMALLINT COMPRESS 1 , Mapping_Id CHAR(10) CHARACTER SET LATIN NOT CASESPECIFIC COMPRESS ('MANUAL','MHT0103008','MHT0103022','MHT0103033','MHT0103034','MHT0103037','MHT0 103064','MHT0103070','MIT0103022','MIT0103033','MIT0103034','MIT0103037','MIT010 3058','MIT0103062','MIT0103064'), Load_Time TIMESTAMP(6)) UNIQUE PRIMARY INDEX XAK1EVENT_GL_AMOUNT ( Event_Id ,Event_Amount_Type_Cd ) PARTITION BY CASE_N( EVENT_AMOUNT_TYPE_CD = 1 , EVENT_AMOUNT_TYPE_CD = 2 , EVENT_AMOUNT_TYPE_CD = 3 , EVENT_AMOUNT_TYPE_CD = 4 , EVENT_AMOUNT_TYPE_CD = 5 , EVENT_AMOUNT_TYPE_CD = 6 , EVENT_AMOUNT_TYPE_CD = 7 , EVENT_AMOUNT_TYPE_CD = 8 ,... this range is upto 280 NO CASE, UNKNOWN);
|Copyright 2016 - All Rights Reserved|
|Last Modified: 27 Dec 2016|