• We should now be fully online following an overnight outage. Apologies for any inconvenience, we do not expect there to be any further issues.

(why) should I wait for Ivy Bridge?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

greenhawk

Platinum Member
Feb 23, 2011
2,007
1
71
reading the OP's opening post, I was wondering what virtual hardware the OP was running. If using large ram/multiple cpu images, then having several of them over lapping the host hardware is just asking for trouble. With the need for multiple VM's and using a quad + 8GB, then sticking to VM's of single core 512MB would be advisable if you need several running at once.

Of course, if replicating a real world setup, that flexability goes out the door and so you really need to look at spending the money on a dual cpu system just to get enough cores to work with (and by extention, REG/ECC memory to get the large memory needs as well).
 
Jul 10, 2007
12,041
3
0
reading the OP's opening post, I was wondering what virtual hardware the OP was running. If using large ram/multiple cpu images, then having several of them over lapping the host hardware is just asking for trouble. With the need for multiple VM's and using a quad + 8GB, then sticking to VM's of single core 512MB would be advisable if you need several running at once.

Of course, if replicating a real world setup, that flexability goes out the door and so you really need to look at spending the money on a dual cpu system just to get enough cores to work with (and by extention, REG/ECC memory to get the large memory needs as well).

DC has 512mb ram, 1 cpu.
web servers, 1gb ram, 1 cpu.
SQL boxes have 1gb ram, 2 cpu's (underpowered IMO).
 

greenhawk

Platinum Member
Feb 23, 2011
2,007
1
71
DC has 512mb ram, 1 cpu.
web servers, 1gb ram, 1 cpu.
SQL boxes have 1gb ram, 2 cpu's (underpowered IMO).


so should be well inside the ability of the current system then.

there goes that recommendation to get to several VM's.

Drive load spreading might be good for the DC/SQL VM's, but that only goes so far.

I've not played heavily with VM's for a while, but out of interest, none of your drives are green 4K ones that are storing the VM's drive images? parition miss-alinement between the actual and the VM's might cause extra unneeded load on the system.

Enough to warrent the system grinding to a halt is another thing all together. Though the same issues could occur if the VM disks are on a raid 5 array as well.