At 12:57 AM 10/11/2003 +0000, you wrote:

> > > I have a table called history with the below attributes:
> > > IPAddress int ,
> > > RowId int ,
> > > DateTime timestamp ,
> > > RawSec int ,
> > > RawUSec int ,
> > > SeqSec int ,
> > > SeqUSec int ,
> > > CurrValue float ,
> > > PresentState smallint ,
> > > DataQuality int ,
> > > FlagManuallySet char(1) not null,
> > > FlagBlocked char(1) not null,
> > > FlagControlInhibit char(1) not null,
> > > FlagTelemFailed char(1) not null,
> > > FlagTestMode char(1) not null,
> > > FlagCalcFailure char(1) not null,
> > > FlagLinkFail char(1) not null,
> > > FlagPointFault char(1) not null,
> > > FlagUnreasonable char(1) not null
> > >
> > > Currently this table has about 22 million of records. The index
>has
> > > been created on IPAddress, Rowid and DateTime. When I select from
> > > this table, I encountered performance problem (as user complained
> > > about it). I have no idea how to improve the performance. Can
>anyone
> > > help ?
>
>Can you advise me what are the information you need ?
>
>I always select with condition such as "where IPAddress = 123 and
>RowId = 345 and DateTime >= '2003-10-1 00:00:00' and DateTime < '2003-
>10-2 00:00:00'". and I create index key on IPAddress, RowId and
>DateTime.
>
>Is the above information help ?

Getting there.

What is the primary key of this table?

If the primary key has the same segments as your index, do you therefore
have *both* a primary and an index on this group?

What kind of performance problem does your user complain about?

What does the query plan look like?

Do you have any other indexes on this table?

heLen