Thanks to all for the responses. Further info:
1. I have researched the various config changes you all discussed at length - thanks for confirming my findings :) The DC/GC platform will only be used for DC/GC functions - it will not host e2k3 or any other app, so (I believe) the /3gb *is* appropriate and worthwhile. 2. I should have given OS and environmental details: 5Gb DIT; 250,000 objects; w2k3 DC in a native w2k native domain. 3. I have used adperf and adtest in the past but they are IMHO cumbersome at best. SPA I would suggest, is a t.shooting tool. I'm looking for something which can help simulate large client loads (LDAP/Kerb/GAL etc) and which will help me understand where bottlenecks exists in each h/w configuration tested. I guess there is no such 'off the shelf' tool, rather, I shall have to use perfmon and the services of MCS :) Thanks again for the great feedback, neil -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Gil Kirkpatrick Sent: 06 December 2004 17:51 To: [EMAIL PROTECTED] Subject: RE: [ActiveDir] Stress testing and performance analysis of domain controllers See http://www.microsoft.com/resources/documentation/WindowsServ/2003/all/de ployguide/en-us/dssbj_dcc_imef.asp for more MSFT-approved information re: /3gb on DCs. Server Performance Advisor http://www.microsoft.com/downloads/details.aspx?FamilyID=61a41d78-e4aa-4 7b9-901b-cf85da075a73&displaylang=en might provide some insight but it is not predictive. There was a company in the UK named NTSim that had a performance-simulation program for AD, but I think they may have gone under. I've used the ADTEST programs and scripts from Microsoft to generate repeatable loads and it seems to work well, even if it is a bit of a pain to set up. You didn't say anything about the size of your DIT, but generally, providing enough RAM to cache the entire DIT plus indices is a big win. Several large customers have deployed 64-bit DCs with gobs (1 Gob = 8GB :) of memory to do this and have been quite pleased. -gil Gil Kirkpatrick CTO, NetPro "To fly, flip away backhanded. Flat flip flies straight. Tilted flip curves. Experiment!" -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Brett Shirley Sent: Monday, December 06, 2004 10:34 AM To: [EMAIL PROTECTED] Subject: RE: [ActiveDir] Stress testing and performance analysis of domain controllers Really? Z:\ntds\db>dir ... 05/20/2004 07:47 AM 7,899,987,968 ntds.dit ... Cheers, -BrettSh On Mon, 6 Dec 2004, Renouf, Phil wrote: > You don't need the /3GB switch for a DC. Just having more than 2GB of > ram does not require using the /3GB switch, systems like Exchange > require it, but a DC shouldn't need it. > > Phil > > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of Ruston, Neil > Sent: Monday, December 06, 2004 11:57 AM > To: '[EMAIL PROTECTED]' > Subject: [ActiveDir] Stress testing and performance analysis of domain > controllers > > As part of a more general AD design refresh, I am re-visiting the DC > hardware and OS configuration. > > I am proposing several changes to the DC spec, including the adoption of > the following: > > * Use 4Gb RAM > * Use /3gb switch > * Place AD logs and database on separate disk spindles > > > In order to 'sell' this idea, I would like to demonstrate the effective > increase in 'horse power' that the above offers. I am therefore looking > for a tool which can help me to show that a DC with config A can handle > load x whilst DC spec B can handle load y. > > Ideally, this tool will act much like loadsim and simulate a load on the > DC so as to identify the maximum load that each config is capable of > handling. > > Is there such a tool available on the market? > > Thanks in advance, > Neil > > Neil Ruston - MVP Directory Services > > ======================================================================== > ====== > This message is for the sole use of the intended recipient. If you > received this message in error please delete it and notify us. If this > message was misdirected, CSFB does not waive any confidentiality or > privilege. CSFB retains and monitors electronic communications sent > through its network. Instructions transmitted over this system are not > binding on CSFB until they are confirmed by us. Message transmission is > not guaranteed to be secure. > ======================================================================== > ====== > > > List info : http://www.activedir.org/mail_list.htm > List FAQ : http://www.activedir.org/list_faq.htm > List archive: http://www.mail-archive.com/activedir%40mail.activedir.org/ > List info : http://www.activedir.org/mail_list.htm List FAQ : http://www.activedir.org/list_faq.htm List archive: http://www.mail-archive.com/activedir%40mail.activedir.org/ List info : http://www.activedir.org/mail_list.htm List FAQ : http://www.activedir.org/list_faq.htm List archive: http://www.mail-archive.com/activedir%40mail.activedir.org/ ============================================================================== This message is for the sole use of the intended recipient. If you received this message in error please delete it and notify us. If this message was misdirected, CSFB does not waive any confidentiality or privilege. CSFB retains and monitors electronic communications sent through its network. Instructions transmitted over this system are not binding on CSFB until they are confirmed by us. Message transmission is not guaranteed to be secure. ============================================================================== List info : http://www.activedir.org/mail_list.htm List FAQ : http://www.activedir.org/list_faq.htm List archive: http://www.mail-archive.com/activedir%40mail.activedir.org/
