To improve the I/O performance for ESX Virtual Infrastructure, VMware had come out with the round robin option for both ESX and VM guests. Although is an experimental option in the ESX setting today, but I will encourage you all to try this option which provide fail over and load balancing on the storage path to connect to you SAN storage. For VM guests, you will allow allow to use this when you have RDM – Raw Device Mapping option to direct read write to the physical LUN from your SAN storage without using VMFS.

To enable this on ESX host, you need to browse to the configuration tab of the ESX host, and right click the data store and select properties, and click on manage paths option in the GUI wizard. Click on Change button after that, and choose the Round Robin (Experimental) option and click OK. You will need to go through this process 1 by 1 to ensure you had round robin from each ESX host to each of the VMFS Data store.

For VM Guests, Just right click the VM and choose edit setting, and select to the hard disk which has shown as Mapped Raw LUN on the summary tab. Click on the Manage Paths and follow by the change button, and same you can easily configure to have the Round Robin enable.

ESX Host Configuuration

Wizard View for each Data Store connection from ESX

Policy Option which allow to change for Round Robin

Select the option of Round Robin

VM Guests Configuration

Click on Manage Paths to continue with the Round Robin Option in the next screen

You will actually see the same display and option at the last 2 steps compare the VM guests and ESX Host

Recently I had read a couple of article regarding the performance caparison chart from VMware, Netapps and some of the forum communities, I do really find out the real performance is much different with the technical white paper that I read before this.

As for the today, more users are actually deployed the mission critical and high I/O servers on the virtualization environment, but we do see some I/O bottle neck which cause by the storage performance always. VMDK do provide flexibility from management perspective, but it does sacrifice the performance you may require for your databases, files transfers and disk performance. I had run a couple of test with real case scenerio instead of I/O meter that been always use widely, and here is the summarize result I would like to share.

In disk perfomance, we always split it to 2 categories as sequential and random I/O. in sequential mode, you will see the huge different while you try to perform the file transfer locally or through network. My test environment is running with SAN storage from fiber channel with same LUN size and raid group which are created from the Storage Level. The only differences is VMFS Vs Raw.

Raid Group design 7+1 raid 5 configuration and run on MetaLun configuration

Each LUN size is 300GB

Performance monitoring tools = Virtual Center Performance Chart

VM Test Machine = 4 Vcpu, 8GB Memory

Operating System = SLES 10 x32, x64 ; Windows Server 2003 x32, x64

Sequential : RDM is out perform VS VMFS as it able to achieve > 2 times higher through put during the file transfer locally on the VM

Random I/O : The Raw Device Mapping is still out perform the VMFS and getting the similar through put with sequential file transfer. Multi session with random database query is been executed in the test

for NFS file transfer from VMFS to VMFS, I do see the bottle neck happen much more earlier than RDM.

Highest I/O rate for RDM = 180MB/s as during sequential files copy and DB query

Highest I/O rate for VMDK = 100MB/s during sequential files copy and DB query