Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Usenet -> c.d.o.misc -> Re: Simple Oracle Query Taking Too Long
Bryan Hunter wrote:
> The table that I am migrating the data to is not relivent to the problem, so
> I did not feel that the number of rows or indexing was important just the
> process that I was performing.
This may well be your feeling but we are, it would seem, disagreeing.
> I did copy a portion of the data to a new table as suggested with no
> indexing and selecting where rownum = 1, still took almost 1 minute.
What was Oracle doing during that minute? Do you know?
> The original table has initial extent of 64kb and a 1% increase with
> unlimited extents
That is about as horrible a definition as I can imagine. I can't think of a single justification for either the 64K extents or a pctincrease that is anything except 1.
And if you are in 8.1.7 then I can't think of any justification for dictionary managed tablespaces either ... not that any of this can explain your current issue.
> I do not know if I agree with the statement that select the first 1000 rows
> of a table will result in a different 100 0rows every time as suggested by G
> Quesnel. I would appreciate any comments on this also.
Your agreement is not required. The statement is correct as long as DML is taking place. ;-)
How long does it take to run your SELECT statement against other tables of similar size?
-- Daniel Morgan University of Washington Puget Sound Oracle Users GroupReceived on Wed Sep 06 2006 - 18:20:23 CDT
![]() |
![]() |