|
Home > Archive > Microsoft SQL Server forum > June 2005 > Win 2003 x64 + SQL 2005 < Win 2003 32 + SQL 2000?
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Win 2003 x64 + SQL 2005 < Win 2003 32 + SQL 2000?
|
|
| rcamarda 2005-06-23, 7:23 am |
| I may not be the best forum for this post, however I believe some one
can still help me or direct me somewhere else.
I'm finding that win 2003 x64 and SQL 2005 CTP is slower than win 2003
and SQL 2000.
Machine A: Opteron 2.2 (248) w/ 2 cpus. 8GB memory. SCSI disk array.
Windows Enterprise 2003 x64 final release. SQL Server 2005 enterprise
CTP.
Machine B: Opeteron 2.2 (248) w/ 2 CPUs. 4GB Memory. IDE Drives.
Windows Server 2003 (32bit) and SQL Server 2005 (standard).
I'm using Cognos' DecisionStream as the ETL tool (32bit). I have the
identical job on both machines. Machine A is slower to do my nightly
build (1h 20m) vs. Machine B (50m).
I've done no tweaking of the databases. I Used SQL 2005's studio to
copy the database from Machine B to Machine A.
The only major difference is the O/S and SQL, the machines are same
speed, however Machine A has twice the memory and faster disk; so I
would expect it to be faster.
Can anyone think of smoking guns I might be missing?
TIA
| |
|
| I highly recommend that you check out Kevin Kline's article on Bare
Metal Tuning in this month's SQL Server Magazine. It does require a
subscription, but he goes through a series of steps to show you what
changes to hardware configuration will do to SQL Server performance.
One thing he noted in a live presentation of this at DevTeach is that
when you do these tests, you need to be careful about changing multiple
things in between benchmarks. For example, you said the only major
difference is the O/S and SQL Type, but yet your hardware specs are
very different.
Questions that come to mind for me are:
1. Do you have AWE enabled on both machines?
2. Is hyperthreading on or off on both machines (is an Opteron even an
hyperthreaded CPU?)
3. Are your database objects segregated in the same fashion on both
machines?
Again, I really recommend Kevin's article; it might give you some leads
in how to explain the differences.
HTH,
Stu
| |
| Erland Sommarskog 2005-06-23, 8:23 pm |
| rcamarda (rcamarda@cablespeed
.com) writes:
> I may not be the best forum for this post, however I believe some one
> can still help me or direct me somewhere else.
> I'm finding that win 2003 x64 and SQL 2005 CTP is slower than win 2003
> and SQL 2000.
> Machine A: Opteron 2.2 (248) w/ 2 cpus. 8GB memory. SCSI disk array.
> Windows Enterprise 2003 x64 final release. SQL Server 2005 enterprise
> CTP.
>
> Machine B: Opeteron 2.2 (248) w/ 2 CPUs. 4GB Memory. IDE Drives.
> Windows Server 2003 (32bit) and SQL Server 2005 (standard).
Certainly the best group to post to would be
microsoft.private.sqlserver20005.performance.
You find access information on http://go.microsoft.com/fwlink/?linkid=31765.
--
Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se
Books Online for SQL Server SP3 at
http://www.microsoft.com/sql/techin.../2000/books.asp
|
|
|
|
|