Hi,
I'm running an all-in-one Exchange 2013 server. This is supporting around 60 users who work in online mode with larger-than-average mailbox sizes.
The machine is running on ESX 5.1 and is spec'd with the following
4 x 2Ghz
16GB memory
Disks on dedicated LUN on RAID10, 15K shared storage with 8GB FC.
Performance at desktops appears fine. However, I have been watching performance monitor and some stats look particularly high. For example (all using default scales):
Processor - Interrupts/Sec (Average) = 2,000 (min 1,150 - max 5,900)
Processor - % Processor time (Average) = 9 (min 0.005 - max 53.9)
Memory - Page Faults/sec (Average) = 1,300 (min 114 - max 13,800)
Memory - Page Reads/sec (Average) = 0.3 (min 0 - max 2)
Memory - Pool Paged Bytes (Average) = 379,240,000 (min 379,000,000 - max 379,322,000)
Memory - Pool NonPaged Bytes (Average) = 142,300,000 (min 142,180,000 - max 142,600,000)
Memory - Cache Faults/Sec (Average) = 34 (min 1 - max 446)
My concern is around the memory, particularly the page faults. Should I be concerned? if so, what is the best way to address this considering it's a VM. There is more memory in the host available, but I don't want to just throw memory at the problem if it will
just mask the issue.
Comments/suggestions greatly appreciated.