Re: separate data/inidex

Mike,

How many discs do you have in your laptop ? (Can I assume just one)

Do you use your laptop as a dedicated
Oracle server with no other conflicting
resource demands on the disc
(Clearly not, as you point out, you are doing lots of examples all the time, and no doubt documenting it, and as you point out "running other applications").

How much space does NT allocate for swap during the course of the day on that one disc. and how much impact does that have on your processing ? (I upgraded my childrens' PC with 128MB of RAM a little while ago, and do you know what goes faster ? Shutting down -I think because the shutdown command no longer has to re-read the swap to find out that it doesn't have any tidying up to do).

I'm not arguing that you are wrong, of course, but I don't really think that you can quote your laptop as a valid example on this one - there are too many other factors involved.

--
Jonathan Lewis
http://www.jlcomp.demon.co.uk

Author of:
Practical Oracle 8i: Building Efficient Databases

Next Seminar - Australia - July/August
http://www.jlcomp.demon.co.uk/seminar.html

Host to The Co-Operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html



Mike Ault wrote in message
<37fab3ab.0205050725.56ca9efc_at_posting.google.com>...


>

>It will create the datafile in pieces all over the disk, not as a

>contiguous area. Any subsequent extensions will be created in a

>similar manner. Afraid we are back to the old VMS as far as NT

>(gee...since many of the VMS staff worked on NT do you thing there is

>a parallel?) defrag to improve performance. An example is that on my

>laptop I had been merrily going along adding, deleteing, creating

>datafiles, dropping them...then I defragged...it took seven runs

>through the defrag routine befor it was defragmented...even in areas

>where supposedly nothing had been written. Performance for all

>applicaitons including Oracle improved. Afraid it has reached that

>point once again, the drive light stays on more than the CPU activity

>light.

>

>Yes, if you system is defragmented when you create the tablespaces and

>nothing is done to the disk to cause additional fragmentation, then it

>shouldn't be required, I will add to my comment, monitor and defrag

>the disks as needed.

>

>Mike