Re: Export times: affected by size of table or amount of data?

In article <929459472.25438.0.nnrp-01.9e984b29_at_news.demon.co.uk>, Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk> writes
>
>export time = read time + write time.
>
>Is it possible that your customer now has an
>'empty' table that has actually had its high water
>mark pushed up to the 100Mb mark, so that
>the read time (tablescan) of the export is high
>even though there is no data to write ?
>

Coo... ah... I wonder... Looks like I'll have to find the documentation!

Andy
>--
>
>Jonathan Lewis
>Yet another Oracle-related web site: www.jlcomp.demon.co.uk
>
>TurkBear wrote in message <376666ca.1629533_at_news.dot.state.mn.us>...
>
>>Andy Hardy <aph_at_ahardy.demon.co.uk> wrote:
>>>If I declare a 100Mb table with no data, will it export in less time
>>>than the same table full of data?
>>>
>>>I thought that export times were only dependent on the quantity of data
>>>stored, but my customer is telling me otherwise...
>>
>>Unless I am misreading my logs, the time required for export is dependent
>on the
>>number of rows ( amount of data ) exported plus the various grants, indexes
>and
>>constraints ( if any ) on that table....
>>
>>So your empty table, regardless of size, should be much faster than one of
>the
>>same # of MBs ( allocated ) with, say, 250000 rows of actual data...
>>
>
>
>

--
Andy Hardy. PGP key available on request