Manage Learn to apply best practices and optimize your operations.

Calculating virtualization ROI using unconventional factors

Calculating ROI for a virtual data center is a tricky endeavor. For a true assessment, you’ll need to include several, unconventional metrics.

Accurately calculating ROI for a virtual data center involves a variety of metrics -- one of which is not necessarily reduced hardware costs, contrary to conventional wisdom.

One of the biggest justifications for server virtualization has always been that it reduces hardware costs. But most organizations find the opposite is true: Virtual machine sprawl has driven up hardware expenditures.

As such, you’ll need other metrics to demonstrate a virtual data center’s return on investment (ROI), including savings from high availability, licensing costs and simplified provisioning.

Calculating ROI with high-availability metrics
When an application server drops offline, there are real costs associated with the outage. Depending on the application, these costs may come in the form of reduced user productivity or lost online sales. In any case, the failure of an application server will affect the organization’s bottom line.

Unfortunately, traditional high-availability tools (e.g., failover clustering) aren’t always an option. For example, you can only use failover clustering if an application is cluster aware. This requirement isn’t a problem for many higher-end, enterprise-class applications, but many legacy applications aren’t cluster aware and are thus left unprotected.

In a virtual data center, you can create a cluster at the hypervisor level to address this problem. In this setup, the virtual servers receive the benefits of clustering, even without cluster-aware applications. Furthermore, live migration technologies – such as VMware vMotion and Microsoft Live Migration -- can move a running virtual machine to different cluster node without any downtime. That way, you can take down and perform maintenance on cluster nodes on an as-needed basis.

If you use a hypervisor-based high-availability tool, you can calculate the ROI of that investment. First, determine how much a particular application outage has cost your organization, and divide that figure by the number of hours the application was offline – which will give you a rate. Then, multiply the rate by the total hours of downtime that your application experienced last year. The resulting figure is how much downtime for a specific application has cost your company.

Considering that your high-availability solution prevents downtime, you can add that figure to your virtual data center ROI calculation.

Calculating the ROI of reduced licensing costs
You can use licensing cost savings when calculating ROI. For the sake of this discussion, let’s assume that all of your servers run Windows Server 2008 R2.

In a physical data center, Microsoft requires a server license for each server. But in a virtual data center, the rules change. Every copy of Windows Server 2008 R2 Enterprise edition allows for up to four virtual servers on a physical machine, assuming that you don’t run any applications on the host operating system (which would violate Microsoft’s licensing terms). If you opt for the Datacenter edition, this license covers the physical server, and you can run an unlimited number of virtual machines on server.

When embarking on a new IT project involving multiple Windows servers, it’s generally less expensive to deploy new servers as virtual machines. Hence, you can consider the savings on licenses as an immediate ROI.

Quicker provisioning times
When calculating ROI, you can include the IT staff’s increased productivity. Here’s how to measure it:

Suppose an organization frequently deploys new servers for various business requirements. Now pretend that the person who provisions the new servers is paid $35 an hour, and it takes roughly an hour and a half to manually provision a new server. Given these figures (which I am pulling out of thin air), it costs the fictitious organization $52.50 in labor to deploy a new server.

In a virtual data center, servers are usually deployed from templates or cloned from existing virtual servers. Because the administration time decreases, it may take 10 minutes to deploy a new virtual machine. If we use the figures from the previous example, the deployment-related labor costs drop to $5.83 per virtual machine, resulting in a savings and ROI of $46.67.

As you can see, you can use a number of metrics when calculating ROI. The trick is determining your organization’s actual costs and measuring how those figures change once you implement a virtual data center.

Brien M. Posey, MCSE, is a Microsoft Most Valuable Professional for his work with Windows 2000 Server, Exchange Server and IIS. He has served as CIO for a nationwide chain of hospitals and was once in charge of IT security for Fort Knox. He writes regularly for TechTarget sites.

Dig Deeper on Downtime and data loss in virtualized environments

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.