New
#21
Hi there
@Steve C
Since we all like doing Bonkers tests of our systems for doing stuff they were never intended to do :
Linux Host --->lvl 1 : esxi server VM ---->lvl 2 : Windows XP VM ---->lvl3 :windows 10 X64 Pro VM
For fun I --you can do it believe it or not run a 64 bit OS on a 32 bit Host (in this case the Host is XP VM system for W10) - but of course you can only assign 4GB of RAM to Host + guest together !!!!
The W10 system ran but very slowly -- only to be expected but I'm surprised it even ran at all.
Rig used was a 32 GB RAM HP proliant Gen 10 Microserver with AMD Opeteron X4321 2.1 Mhz 4 core processor).
VM's stored on 512GB standard samsung SSD.
Actually I think the main problem with Nesting VM's isn't per se the CPU processing power available but probably the bus speeds on the mobo for transmitting cpu instructions and data and the probable nesting of multiple level of I/O requests -- we aren't speaking here of running Concurrent VM's but nesting them which is a totally different ballgame -- also going through that level of Virtual Network adapters will slow the thing down to crawling speed. --
Still a nice interesting time wasting exercise -- given better bus speeds etc I'm sure you could actually get much more significant throughput on nested levels of VM's.
BTW as standard VM's i.e just running on the Host i can quite easily get around 10 VM's to run concurrently -- however I've really got nothing to stress test that -- what one needs is some sort of background test set for each VM with maybe a user or 2 logged on to each of the VM's. ==Any ideas ???
Cheers
jimbo
Quote
