> Hi
>
> If you use profiler to see what SCD does, you will notice that it is
> not the
> best way to do it when the number
> You can implement your own SCD, using merge joints, OLEDB commands,
> and set
> updates to speed up your process.
> cheers
>
> "Scott" <Scott@discussions.microsoft.com> wrote in message
> news:A7E07F71-DACB-4D06-8013-7CEC98F242A6@microsoft.com...
>
>> I am using an SCD transform with a dimension table with 450,000
>> records.
>> The
>> SCD is really slow. Does anyone know anything I may be able to do to
>> speed
>> things up?
>> Thanks for the help!
>> Scott

"Allan Mitchell" <allan@no-spam.sqldts.com> wrote in message
news:b067a6253e628c971e3db688d84@news.microsoft.com...
> Hello Francisco,
>
>
> I agree with looking at the output from Profiler but I would also look to
> use this info to maybe explore the option of building indexes in the dim
> tables to optimise the queries.
>
> If perf is a problem then I would err on not using OLE DB CMD Transforms
> as they are a Row*Row operation and a large number of rows passing through
> it == painful.
>
> I would be inclined to stage and do a SET based update/delete if perf
> really is that bad.
>
>
>
> --
>
> Allan Mitchell
> http://wiki.sqlis.com | http://www.sqlis.com | http://www.sqldts.com |
> http://www.konesans.com
>
>> Hi
>>
>> If you use profiler to see what SCD does, you will notice that it is
>> not the
>> best way to do it when the number
>> You can implement your own SCD, using merge joints, OLEDB commands,
>> and set
>> updates to speed up your process.
>> cheers
>>
>> "Scott" <Scott@discussions.microsoft.com> wrote in message
>> news:A7E07F71-DACB-4D06-8013-7CEC98F242A6@microsoft.com...
>>
>>> I am using an SCD transform with a dimension table with 450,000
>>> records.
>>> The
>>> SCD is really slow. Does anyone know anything I may be able to do to
>>> speed
>>> things up?
>>> Thanks for the help!
>>> Scott
>
>