If you have assigned more memory to the guests than the host has physical RAM or if you have applied a memory restriction to a guest ESX has some mechanisms to server memory requirements if a guest needs more memory, which is not available.
The first thing it does is it expands the balloon driver in the memory one guest, forcing that one to swap out to its disk (guest swapping) because the balloon driver requests RAM. That RAM is then cut off from this guest and granted to another guest which has requested memory to use it actively.
If that is not enough ESX starts to emulate RAM for individual guests, but swapping out guest memory to ESX datastores (host swapping).
Both situation shouldn't occur if you have more or at least an equal amount of memeory as you have assigned to all guests in total.
Both situations slow down the processing.
AWo
VCP 3 & 4
\[:o]===\[o:]
=Would you like to have this posting as a ringtone on your cell phone?=
=Send "Posting" to 911 for only $999999,99!=