Building the business case for a server refresh

As IT managers consider a server refresh, they need to create a solid business case for replacing hardware. Here’s how to garner support for the project.

This article can also be found in the Premium Editorial Download: IT in Europe: Server hardware and refresh cycle options

Editor’s note: This is the second in a series of four articles on server hardware equipment refreshes. Part one detailed strategies for getting the timing right; part two shows you how to garner support for the project.

Server refresh strategies can be difficult to justify to business stakeholders. And when your existing servers were approved only three or four years ago, building a solid business case for new purchases gets even more difficult. Another challenge is gaining trust among stakeholders, especially when the perception is that a server refresh strategy has been proposed purely for technology’s sake and not to benefit the business.

When building a business case, consider the following factors.

Operational cost reductions

Vendor-touted power and space efficiencies from investing in newer-generation servers become key points to focus on within a business case proposal, especially in today’s world of ever increasing power costs and real estate margins.

Justify those points using the following examples:

  • Data centre and real estate costs
  • Heating, ventilation and air conditioning (HVAC) costs
  • Total data centre power consumption
  • Connectivity costs
  • Power consumption of older inefficient servers
  • Management tools and people costs

Ensure that any vendor-touted benefits reduce your company’s specific operational cost burdens. This level of due diligence demonstrates to stakeholders that a server refresh can reduce costs elsewhere in the business and that the proposed server refresh initiative is not as big a burden as capital expenditure numbers might indicate on their own.

Blade servers can reduce the total consumed physical rack footprint and total connectivity to SAN/LAN devices. And some models enable better power/HVAC efficiency. At the review stage, consider the total costs and up-front investment of migrating to a blade architecture.

To demonstrate more tangible cost savings, collaborate with building facilities management teams. They will likely know how much money can be saved by reducing operational overhead such as facility space and HVAC and provide proof that server consolidation can translate into real-world savings.

By demonstrating potential areas of cost reduction, you can better persuade stakeholders to purchase new hardware. It also gives executives peace of mind knowing that the proposed server refresh is about cost savings, not just investing in the latest and greatest server technology.

Maintenance cost avoidance

Any server eventually reaches end of life, and waiting until EOL puts infrastructure teams into a risky situation. During that time, a business depends on unsupported hardware that lacks readily available or any replacement components.

As an EOL milestone date approaches, research whether a new server refresh outweighs the potential investment in extended or third-party maintenance support. Check whether the support warranty extension is more cost-effective over the long term than buying new server hardware on capital expenditure would be. 

Service reliability

In the current round-the-clock service-availability climate, businesses increasingly rely on IT to host applications and services on a solid, reliable IT infrastructure. But greater demand for application availability in turn puts greater demand on the physical server platform. Servers may have been purchased when a given application was less critical or in usage level by third parties, particularly in cases when the original hardware investment was made before other key shifts in legacy business processes or applications.

Breaking the vicious cycle

IT can become too accommodating in such situations. Here IT may try to satisfy customer requirements without getting the necessary business-side investment in computing resources. This puts IT in the position of doing hat tricks to deliver highly available, robust systems on platforms that receive zero investment. Even worse, the business side probably doesn’t understand the potential risk of delivering these services on aging server infrastructure.

As you build a business case, construct a what-if cost analysis to calculate—in terms of lost business—the cost associated with hardware underinvestment. Consider the impact of a service outage among key applications, such as how much a day of downtime would cost for a business process such as an invoice-processing application or loss of service at a critical business peak period for an external customer-facing e-commerce portal.

For internal, line-of-business-based applications and core services such as email, it may not be easy to calculate such cost. So consider building an engagement process with application-centric IT staff members who know the cost of an outage. You can also create a solid relationship with business stakeholders to gain a better understanding of the cost associated with a service outage.

Last and most important: Educate the business on why it’s critical to invest in new hardware to break the vicious cycle.

Better server consolidation ratios

Many organisations now refresh hardware to gain the consolidation benefits from collaboration between CPU manufacturers and virtualisation companies. These benefits offer the opportunity to virtualise more resource-intensive workloads that were previously unsuitable for virtualised environments. Consolidation ratios are also limited by older chipsets and often require additional physical hosts to continue virtualising.

When proposing a server refresh business case, include the financial benefits of higher virtualisation ratios. Highlight to stakeholders that this continued, aggressive consolidation creates further cost benefits, such as reduction in data centre space, operational managerial overhead, support and maintenance, and so on. But note, too, that better consolidation ratios place greater demands on host hardware availability, which will need to be designed to be more robust to handle an increasing number of virtual machines.

Application licensing

A financial investment in newer-generation servers which have better performant processors within a smaller total physical socket count—can reduce the cost of purchasing licenses and maintenance renewals. Some independent software vendors have embraced multicore servers, especially because licensing is often based on licensing per physical CPU socket.

As they create business case proposals, data centre managers should review current licensing expenditures and inventory to assess whether investing in more powerful multicore servers could reduce total CPU socket licensing requirements or improve productivity elsewhere. Additionally, if your business case doesn’t get approval, use the opportunity to learn about potential cost reductions with multicore servers for databases or middleware server building blocks.

Daniel Eason is a UK-based infrastructure architect at a multinational company. He also has a personal technology blog: http://www.vmlover.com.

Read more on Managing servers and operating systems

Data Center
Data Management