VM /ETC

vSphere has introduced several new features for storage performance enhancement. Most of the new features build on already accepted vSwitch standards and designs. An important example is the new Round Robin MPIO path policy for VMFS LUNs. However, based on what is the common vSwitch design today, the new iSCSI configuration needed for Round Robin multi-pathing may cause some admins to look twice.

I was motivated to write this post by 2 recently published storage vendor documents that both recommend the same basic iSCSI vSwitch with Round Robin MPIO configuration: create a single iSCSI vSwitch, assign 2 physical NICs, and then create as many as 8 VMKernel Portgroups each with their own ip address. The documents I am referring to are:

• NetApp TR-3749 – NetApp and VMware vSphere Storage Best Practices http://media.netapp.com/documents/tr-3749.pdf
• Dell Equallogic CONFIGURING VMWARE VSPHERE SOFTWARE ISCSI WITH DELL EQUALLOGIC PS SERIES STORAGE – http://www.equallogic.com/resourcecenter/assetview.aspx?id=8453

To give a visual of the recommended configuration (in case you are still doing a double take) here are screen shots of the configured vSwitch from:

NetApp’s PDF

Dell’s PDF

Read the document(s) for the specific steps, requirements, and options to configure each vendor’s iSCSI storage with ESX 4 Round Robin multi-pathing, but the reoccurring process found in both documents goes like this (modeled after the table of contents of the Dell PDF):

Configure vSwitch and Enable Jumbo Frames
Add iSCSI VMkernel Ports
Assign 2 physical NICs to iSCSI vSwitch (at least)
Associate VMkernel Ports to physical NICs and override load balancing policy with dedicated pNICs
Enable VMware iSCSI Software Initiator
Bind VMkernel Ports to iSCSI Software Initiator
Connect ESX to iSCSI Storage
Enabling VMware Native Multipathing – Round Robin
Create VMFS Datastores

Why use multiple iSCSI VKernel Portgroups in vSphere? Vaughn Stewart explains in his blog post vSphere on NetApp Best Practices Released:

“iSCSI has made major gains in vSphere and in TR-3749 we detail how to aggregate the bandwidth of multiple network links. This solution allows customers with GbE based storage networks to achieve {omitted} the throughput needs of their most demanding applications without having to upgrade the storage network to FC, FCOE, or 10 GbE.”

Let me know if there are other whitepapers or technical notes published by iSCSI storage vendors other than Equallogic and NetApp that recommend the same configurations.

Related Posts

• VCP4 Upgrade Deadline Extended to January 31st, 2010. Second Shot Testing Program Extended Also. (5)
• Don’t Hit Ctrl+Alt+Del On the ESX 4 Console (12)
• Real Thin Provisioning And Over Allocation – The VI Admin (28)
• Provision a Thin Provisioned Standby LUN For vSphere Thin Provisioning (13)
• Quick List Of VCP 4 Study Links Posts (16)
• Tap into vSphere PVSCSI Performance with Separate VM Boot and Data Drives (11)
• Recommended Reading (0)
• vSphere 4.0 Best Practices (6)
• Reasons For Using NFS With VMware Virtual Infrastructure (14)
• vSphere Migration Checklist (2)