RE: Find Difference between 2 Tables, 22 GB each, on unique key

Folks

Need to get=A0the difference records between 2 tables each of size 22GB =
(each containing 32 Million rows). The temp tablespace is 30GB.

Unique index exists on the respective fieldname on which to check for =
difference.

Which of the following approaches is faster/most efficient?

1) Select <fieldname> from Table 1 MINUS Select <fieldname> from Table =
2? - If so what would be the consumption of TEMP Tablespace?=20
Additionally how to concurrently monitor consumption of Temp tablespace =
(used/free) while the job is running?

2) using some Other SQL script

3) using a PL/SQL script


Qs Will doing a FULL index ONLY Scan be faster than a FULL Table scan?


Machine configuration -  4 cpu's and 1GB DIMM RAM.=20
OS version - Unix Sun Solaris machine Version not known.
Oracle version -=A0 8.1
Nature of application -=A0Joins and inserts
production issue -=A0 Yes
Tables being accessed by other applications - YES, during certain time =
slot. We have abt 12-14hrs when they are not touched by any application.
=A0
Thanks
--
http://www.freelists.org/webpage/oracle-l

Other related posts: