I run a series of PowerShell tests on each server and database every morning. One is the test-mapiconnectivity cmdlet. The odd thing is that on servers with more than one DB the second and third DBs always report higher latency than the first, every day on each server. All the DBs are on the same SAN over the same fabric; yet the second and third always report higher latency. I am not sure if this is real or some artifact of how the cmdlet measures latency. i run the cmdlet with no options specified. TBrennan

Just because they're on the same SAN doesn't mean that they're on the same disk aggregate.Ed Crowley MVP "There are seldom good technological solutions to behavioral problems."

Hi TBrennan, Could you please post the information here, we could do more research about what you referred. How about use netmon to retrieve some information. Regards! Gavin TechNet Subscriber Support in forum If you have any feedback on our support, please contact tngfb@microsoft.com Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread.

I didn't mean for this forum to try to troubleshoot the latency problem itself; I'm just trying to see if anyone else has seen this pattern in the test-mapiconnectivity cmdlet. These are the results for one server for July: the second store has a higher latency for everyday except one. Another server with three stores shows the same thing: the first store is usually under 20msec the other two always over and always higher than the first store. The SAN admins say all the stores are on the same disk group on the same EVA. The SAN is divided up in classic SAN vendor fashion with everything dumped into one humongous disk group. While I have seen this have an adverse impact on store performance in Exchange 2003 it shouldn't have a different impact on stores simply because one is the "first" store and the other is the "second" store. From what I'm reading in the two replies so far the cmdlet shouldn't be the difference either. 13 17 10 23 9 52 9 28 9 27 9 24 9 64 9 34 9 39 9 39 9 29 10 42 10 29 10 34 8 36 9 68 8 34 9 44 9 29 9 28 42 32 9 31 10 44 9 38 12 42 10 35 10 34 12 36 10 27 9 27 9 39 TBrennan

Hi TBrennan Per the information, all the value measured by ms, right? And I also did some tests in my sencarioes, it do is always different from each other due to different database/different DISK. It is expected, and the difference is so little that no need to care it, I think that we should not just care the vales of the latency, because those are just measured by ms, they so short. :) Regards! Gavin TechNet Subscriber Support in forum If you have any feedback on our support, please contacttngfb@microsoft.com Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread.