What is faster in a virtual environment?

I have two virtual servers. I am running Hyper-V on 2008 Sever R2 on both.

Server 1
IBM x3650 (7979) ( Dual XEONS total 4 cores)
CPUs are Dual Core Xeon 5150 2.66 GHz Socket 771 LGA (4MB Cache, 1333MHz FSB)
30 GB DDR2 PC2-5300 • CL=5 • Fully Buffered • ECC • DDR2-667 • (IBM)

Server 2
Custom Built w/SuperMicro X7DB8 Motherboard (Dual Xeons total 8 cores)
CPUs are Quad Core Xeon E5320 1.86GHz Socket 771 LGA (8MB Cache, 1333MHz FSB)
20 GB DDR2 PC2-5300 • CL=5 • Fully Buffered • ECC • DDR2-667 • (KINGSTON)

Both will be used as virtual servers. Here are the virtual machines:

(low activity)
1 Virtual Backup Domain Controller - Live
1 MS SQL Server 2005 (has old crm and accounting system) - occasional access
1 Terminal Server (2003 Server) - used occasionally to access old crm and accounting)

(high activity)
1 MS SQL Server 2005 (Current accounting and CRM) - used everyday
1 Telephone Server (software controls phone system runs on 2008 Server)
1 Terminal Server (2008 Server R2) - used daily by ouside sales reps
1 E-commerce webserver - used daily but not high traffic 20 orders per day avg

I am able to assign single or multiple processors and adjust memory. But I'm not sure which will give me the best benefit based on these type of machines.

My question is if I sign 1 or 2 processors to a vm, how will I know if the VM actually benefits from two processors? And then which processor is actually faster for a VM?

One has a higher cache and 4 cores but lower GHz speed. The other has 2 cores higher ghz speed but half the cache.

Also I can technically swap the processors in both machines allowing me to give more memory to VMs. Currently the one with 8 cores has only 20GB of Memory. While the 4 core IBM has 30Gb of Memory. I don't think I can't swap the memory because it has IBM modules which might be picky or proprietary.
MEATBALLHEROAsked:
Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

wolfcamelCommented:
do they have the same disk systems -
How many users connect to the main terminal server?
The main terminal server will love more ram, the main sql server will love ram and disk speed
0
wolfcamelCommented:
as to the question - how will I know what is faster?
the good thing about a virtual environment is the ability to move VMs around and to test.
The user experience is always the best test - test the apps you use doing the things that are complained about the most. Often a report is something that will take a lot longer than one transaction - so it can be easier to measure.
0

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
MEATBALLHEROAuthor Commented:
Both will be connected to the same SAN Strorage Device via iSCSI connection on dedicated 1GB Ethernet network. So the VMs are all on the same storage device and not on the local virtual hosts.

Which processor is faster when assigning a single processor? The dual core with the higher 2.66 GHz processor speed but only 4MB of Cache. Or the new Xeon with lower 1.86 GHz speed with the 8MB of cache?
0
SommerblinkCommented:
The two processors are from the same family. In fact, the four-core processor you have is basically the same as the two-core processor you have, except Intel put two of those two-core processors on the same die.

As far as the CPU cache, performance should remain similar, since two cores share 4MB and four cores share 8MB, or on average, each core gets 2MB of cache, so at this point, unless your CPU load is highly threadable, the SQL server may do better on the higher GHz core.

But to be honest, with your storage on a iSCSI network, your bottleneck will be storage bandwidth and latency before anything else.

And as wolfcamel pointed out... with virtualization, testing becomes considerably easier to perform, so there is that too.
0
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Server Hardware

From novice to tech pro — start learning today.