I have a database that we are performing many reads againsts, but only a few
writes (actually, there is a bulk insert of about 22,000 records every
monday and that's about the only time we really ever write anything to it).
I have indexes on the large tables (those with more than ...

The table contains 200+ columns.  I have already some 15 non-clustered
indexes on it.  Does adding one (or few) more indexes affect INSERT and
UPDATES on the table?
TIA

I have some SSIS packages where I count the  number of rows before and after
a loading.
So I execute a command:
select count(*) as RowsBefore from mytable

I dealt with the same question for many years and decided to ask community
and hopefully MS gurus on what is the best disk configuration to support
multiple concurent update accesses.
Question is not that trivial as it may seem. Let say I need to run

We have a database in SQL Server 2000. I made a copy of this
database(mdf,ndf,ldf) and attached it to SQL Server 2005 on the same
machine. When I try one particular SP in both of them, SQL Server 2000
performs almost 2.5 times faster than 2005. The execution plan in 2005

Hi All. I have two DB servers with the same Data one is SQL 2000 and the
other one is a SQL 2005 upgrade from 2000. We ran a Web app against both DB
(Same query). The 2000 Server for some reason returns the values twice faster
than 2005.

hi,
I have a query about designing a database. I have column which can take only
two values (1/0) (ideal data type would be bit) but someone suggested me to
keep it as int as the processor upgrades any data type to int while

Importing a text file via DTS into a newly truncated table with a
simple PK on an INT (not identity) column.
Watching the profiler, first it does the select for the datapump, then
it does a bulk insert.  There are a few light transforms in-between.

Gentlemen,
I am trying to determine how I would accomplish the following.  I have A DB
that is currently in production use that contains OLTP data.  Each of the
major logical structures has a base table (ie PART) that has an identity for

First of all, please excuse me for the long post. The post is long but
hopefully the problem described is fairly simple for somewhat experienced
data modellers.
I am a beginner in database design so I was hoping that maybe you can give

Can SQL do it as :
Select count(Top 2 *) from table1
I only want to return 0, 1, or 2 only.

i have 2 GB RAM in my sql server 2000.it is installed on windows 2003
enterprize edition. I have made 20 databases in the database.some databases
size is more than 6 gb.
my sql server is using RAM more than 1.8 GB.