there is also a litle pitfall with ...

and jh.run_date = convert(int,

- how long is it taking the system to allocate 200GB ?

- did you try to copy a 200Gb file to that logical disk ? And how long did that...

so this may be I/O related.

keep in mind that files are being initialized (=nulled out) at allocation time (unless you specify not to).

What filesize are you trying to allocate ?

how about :

declare

@daysago int

set

@daysago = 7

Select

j.name as

thanks for the feedback.

Indeed the GUI stuff is very slow.

IMO the datapart is greath.

regading you filegroup-slowness ... I didn't encounter this.

Is it with the creation of the filegroup, or of the files within...

did you install MSDE ? use SQL2000.MSDE-KB884525-SP4-x86-ENU.EXE

or did you install a personal edition ? use SQL2000-KB884525-SP4-x86-ENU.EXE

or did you install only client components ? use SQL2000-KB884525-SP4-x86-ENU.EXE

and do you also have Analysis...

- I'd start with a stop/start of sqlserver (to clear the performancerelated dmv's)

- then I'd do the standard maintenance !

   dbcc dbreindex / sp_updatestats / dbcc updateusage ('yourdb')

- Then I'd start...

depending on the amount of data you'd expect ....

- you could setup replication to your needs and sync nightly

- you could use the tablediff.exe (check books online).

        this ms-util generates...

If you use a dateandtime var and a datetime column, use a parameterized query in combination with a datetime parameterobject.

e.g. vbs:

' Handle commandobject

set CmdObj = CreateObject("ADODB.Command")

CmdObj.ActiveConnection = ConnObj

CmdObj.CommandText = "select...

you could set the "audit level" to ALL.

and then import the sql-errorlog at regular basis and filter the sysadm-logins.

you'll have to keep an eye at the sql-errorlog because it...

- A clustered index _contains_ the datapages (leaflevel)  !

- if you create a table on a specific filegroup, if there is no clustering index, the data will reside on the...

capture the query-load and examine it !

You may need to add some indexes, expecialy to support group by, order by operations ! as well as joins !

This way you may...

get the creation of temporary tables out of  the transaction if you can !

So if possible,  create your #tmptbl(s) and then begin tran  ... do your stuff ... commit/rollback and cleanup...

I didn't test this myself, but the doc seems good :

http://www.databasejournal.com/features/mssql/article.php/3467291