| Sherlock, Kevin 2005-10-27, 8:21 am |
| What does optdiag output look like for that table and index? I would suspect
that there are a lot of "deleted" rows that are still pointed to. You would
need to "reorg rebuild" the DOL table itself if there is no clustered index
which you could rebuild.
"Jesus M. Salvo Jr." <noone@noone.org> wrote in message
news:43547b38@forums
-1-dub...
> Jesus M. Salvo Jr. wrote:
>
> ------------------------------------------------------------------------------
--------------------------------------------------------------------------------
--------------------------------------------------------------------------------
-----------------
>
>
> I am trying to find out what does ASE do in the event of updating a column
> that is part of an index. The URL below shows what happens with an insert
> and delete ... but does not mention what happens to a non-clustered index
> in an update on a DOL table:
>
>
http://manuals.sybase.com:80/online...ses/%40Generic_
_BookTextView/ 21362;pt=21362#X[col
or=darkred]
>
>
> Also, I have changed the application so that the initial value is now
> '1753-01-01 00:00:01' instead of '1753-01-01 00:00:00'.
>
> Now the first few minutes, "set statistics io" show very few logical pages
> read. However, as time goes by, the number of logical pages read increases
> in small increments.
>
>
> Table: message_notification
_history scan count 1, logical reads: (regular=91
> apf=0 total=91), physical reads: (regular=0 apf=0 total=0), apf IOs used=0
> Total writes for this command: 0
>
> Table: message_notification
_history scan count 1, logical reads:
> (regular=103 apf=0 total=103), physical reads: (regular=0 apf=0 total=0),
> apf IOs used=0
> Total writes for this command: 0
>
> Table: message_notification
_history scan count 1, logical reads:
> (regular=107 apf=0 total=107), physical reads: (regular=0 apf=0 total=0),
> apf IOs used=0
> Total writes for this command: 0
>
>
>[/color]
|