Re: Parallel Query Option

Hi
to acomplish what you want you shold do the following.

create the table with a very small initial extent (2 blocks) and minextents set to 1.

Then use this command once pr. file:
alter table <table_name> allocate extent size <size> datafile <filename>;

This will preallocate the extents you need. When you load the data it will fill extent 1 before filling extent 2 etc. If you want to have the data evenly spread across the extents, then you should split your data in six portions. And use SQL*Loader the load the portions into each file. You can specify wich file to load data into. If you use parallell load and direct path you could load the data inn parallell (ie six loader jobs simultaneos) making the load extremely fast.

Rgds
Steinar Heggelund

: 1 in D, etc.).
: Why is this? I thought the premise of parallel query was to evenly
: distribute your data across spindles as a multiple of your CPU's (which I
: have 6)?
: This is running rdbms v7.2.2.4 and HPv9.0.4/
: Thanks!
: Ernst Renner
: erenner_at_thehartford.com
Received on Fri Mar 07 1997 - 00:00:00 CST