Re: Oracle-managed RAID-like tablespaces

Svend Jensen <svend.jensen_at_it.dk> wrote:

>It is possible to have Oracle 'spread' an object automatically over
>several data files. An eksample. Create a tablespace, localy managed,
>uniform size (say 1M) data file 'drive_1......' size 2000M.
>Add data file to tablespace 'drive_2.... size 2000M and so on until the
>tablespace got the size needed or better.
>Then create the object (table). If the object size is much larger than
>the uniform size, Oracle will create extents in a round robin fashion on
>all the data files in the tablespace. I believe all objects (small and
>large) will start with first extent/segment on first datafile and
>continue on second datafile, datafile three.... until the requested size
>is reached. This schema is seen on 8.1.7EE on windows 2000 advanced server.

Your description is exactly what I was trying to explain. But even then, the question remains unanswered whether a full table scan would go along one file on one disk from begin to end and only then take the second file on the second disk of the same object and same tablespace; or, alternatively, whether a full table scan would go to the first extent of one file, then to the first extent of a 2nd file on a 2nd disk and so on. In this second fashion, one could achieve a certain parallelism by setting up the extent the right (uniform) size.

So you told me that write operation occur round robin. But what about read operations of a full table scan?

By the way, other answers in this thread do not support your explanation.

I hope some experts will comment on this.

Bye
Rick Received on Sun Jul 28 2002 - 16:06:03 CDT