In my last two posts, I’ve examined how hyperscale cloud solutions enable quantum leaps in efficiency, infrastructure elasticity, and economics, particularly through increased agility and server utilization.
In original research conducted with Mainstay, we found that companies that go hyperscale can capture opex savings of up to 75 percent and capex savings of 59 percent, with return on investment of up to 149 percent for large enterprise datacenter operators over a five-year period.
How hyperscale achieves savings
But let’s dive deeper into how hyperscale actually achieves this savings – starting today with the five primary ways it can cut your opex dramatically.
One note on our methods: Estimates of TCO and business impact vary depending on the existing maturity and operating practices of the enterprise data center being studied. To simplify our analysis, we used third-party benchmarks to estimate the current operating practices of datacenters today.
1. Simplified system administration
Today a single system administrator, on average, can manage about 300 servers in a large datacenter. With hyperscale, the number jumps to 5,000 by implementing a software defined infrastructure and automation. Common system administration tasks such as monitoring, notifications, quota management and resource allocation, ghost server discovery, and provisioning across networking, storage, and compute workloads can now be automated and orchestrated by hyperscale management software.
Further, because datacenters can now be managed as a homogeneous infrastructure, fewer component and vendor experts are required. As a result, our research indicates that large datacenter operators can expect to reduce system asset management tasks and reduce FTE requirements by as much as 94 percent for managing storage, server and networking environments.
2. Physical deployments savings
An innovative racking system, combined with efficiency improvements, can significantly reduce infrastructure physical deployment costs per year. Our analysis showed that these savings could reach as high as 76 percent over traditional physical deployment annual costs, as shown in Figure 2. The savings are driven by advanced optical backplane technology, as well as greater CPU utilization. Further, a high-utilization, component-based architecture changes the traditional approach of re-racking and re-cabling after each systems-replacement cycle (every three to five years).
3. Power and cooling savings
Our research shows that a high-utilization platform will enable datacenters to operate with substantially less physical infrastructure. Based on our estimates, the average datacenter could reduce overall power and cooling costs up to 73 percent.
4. Reduced software licensing and support costs
As organizations boost component utilization and cut back on datacenter hardware — particularly CPUs — the cost of software licensing will fall proportionally, and companies could realize as much as a 32 percent reduction in ongoing virtualization software license costs.
5. Lower real estate costs
In most cases, real estate cost savings will take the form of cost avoidance as companies find they can accommodate growth in a smaller footprint. Our analysis estimates that by switching to a hyperscale infrastructure, the typical enterprise could reduce its data center footprint by up to 50 percent.
Want to further explore the economic benefits of hyperscale? If so, start by exploring this topic further in our report: An Economic Study of the Hyperscale Datacenter.