I'm trying to export a table containing around 7 Lakh records from a schema. The table structure is given below.
Column Name Null? Data Type
NODE_ID N INTEGER
REGION_TYPE_ID N INTEGER
LOCALE_ID N INTEGER
DISPLAY_TEXT Y VARCHAR2 (1023 Char)
PROD_DIFF Y CLOB
FILE_NAME N VARCHAR2 (2000 Char)
LAST_MODIFICATION_TIME N DATE
The problem is that its exporting via conventional path even with direct=y option. Also its has taken more than 24 hours and still its going on. Can anybody help speed up the process. The command I used is given below.
1) Specify the buffer size i.e. buffer=10000000
2) The export will be quicker if its run on the database server (cutting out the network)
3) Make sure the export file isnt created on a disk which has the databases data files on it (reduce contention)
4) If its 10g use expdp instead
5) If not you could use a query to cut down on the result set, then set of multiple exports off in parallel doing different parts of the table.
6) Look at the CLOB storage parameters you may be able to optimise this.