Rawpixel - Fotolia


Set up a SCVMM VIP template to control load balancing

With a virtual IP template, you can control the way load balancing is handled in SCVMM for an individual workload. Follow these easy steps to create one.

System Center Virtual Machine Manager provides administrators with the ability to create a virtual IP template that defines load balancer configuration settings for specific types of network traffic. Setting up such a template results in the creation of a named object that resides in the Fabric workspace.

To create a VIP template, open the SCVMM management console, and go to the Fabric workspace. Next, expand the Networking section within the console tree and then locate the VIP template container. Right-click on this container, and select the Create VIP template command from the shortcut menu. This will cause SCVMM to launch the Load balancer VIP template Wizard.

As you can see in Figure A, the wizard's initial screen requires you to provide a name and an optional description for the template you're creating. You'll also need to provide a virtual IP port and a back-end port.

Provide template name and description.
Figure A. Provide a template name, a virtual IP port and a back-end port.

Click Next, and you will be taken to the wizard's Type screen. This screen asks you to specify the type of load balancer you're using. In doing so, the wizard gives you two different choices. The first option is to select a generic load balancer. You can use this option with any supported load balancer. In most cases, you'll probably find that this is the option you have to use.

The other option is to specify a specific load balancer for the VIP template. By default, SCVMM only has knowledge of one specific load balancer, and that is the Microsoft Network Load Balancing (NLB) service. You can see what this option looks like in Figure B.

Specify a template type.
Figure B. Microsoft Network Load Balancing service is the only load balancer SCVMM is aware of.

Click Next, and you will be taken to the Protocol screen. The options displayed on this screen vary based on the type of load balancer you selected. If you selected the Microsoft NLB service, then the wizard simply asks you to choose TCP, UDP or both, as shown in Figure C.

Choose TCP, UDP or both.
Figure C. Your VIP template can support the TCP protocol, the UDP protocol or both.

If, on the other hand, you selected the generic load-balancing option for the VIP template, then the Protocol screen will ask you to choose whether you want to use HTTP, HTTPS passthrough or HTTPS terminate. The HTTPS passthrough option provides Secure Sockets Layer (SSL) encryption that extends all the way to the load-balanced resources. Conversely, the HTTPS terminate option provides SSL encryption between the client and the network load balancer, but the SSL encryption ends there. There is, however, an option to re-encrypt traffic so SSL encryption can be enabled between the load balancer and the back-end servers. You can see what this option looks like in Figure D.

Choose HTTP, HTTPS passthrough or HTTPS terminate.
Figure D. If you use a generic load balancer, then you can opt to use HTTP or HTTPS with either termination or passthrough.

The next screen you'll see is the Persistence screen. This screen contains a checkbox you can select if you want to enable persistence. If you enable persistence, then the network load balancer will try to make sure clients always connect to the same back-end server. Otherwise, clients could conceivably connect to any of the available servers that exist behind the network load balancer. Although persistence has its place, it isn't always necessary. For example, if a load balancer sits in front of a series of identical web front ends, then it might not matter which front-end server a client connects to.

If you use the Microsoft NLB service for the VIP template, then the Persistence screen is the last screen you'll have to deal with. From that point, you would be taken to a Summary screen where you can click Finish to complete the wizard. If you chose the generic network load-balancer option, however, there are a couple of extra screens to complete. The first of these screens is the Health Monitors screen. Completing the screen is optional, but it gives you the ability to periodically send requests to the load balancer in an effort to verify the load balancer's health.

The other screen available to you, if you opted to use a generic network load balancer, is the Load Balancing screen. This screen allows you to choose the load-balancing method you want to use. A simple drop-down list allows you to choose between Round Robin, Fastest Response Time and Least Connections.

Next Steps

Learn about new features in SCVMM 2016

Apply updates with SCVMM Maintenance Mode

Integrate Windows Server Update Services server with SCVMM

Dig Deeper on Microsoft Hyper-V management