UII BRIEFING REPORT 111 | SEPTEMBER 2023
Briefing Report

Efficient servers hold the key to energy efficient data centers

Tracking and improving data center infrastructure energy efficiency — meaning the data center energy use relative to the total work processed — is critical in helping organizations to contain both the costs and carbon emissions that result from IT infrastructure growth. This report compares the outcomes from seven energy efficiency improvement projects using an experience-based model developed by Uptime Institute. The goal of the report is to provide quantitative guidance for enterprises, along with practical considerations.

KEY POINTS

  • Projects that improve server energy efficiency can dramatically lower energy consumption — down to less than half in some cases — while also resulting in equivalent facility power and space reductions. Alternatively, upgrading to more efficient servers can vastly increase processing capacity, by up to 70% in the models, in the same space and power envelope.
  • A 2-to-1 server upgrade from Intel 2019 servers to Intel 2021 servers can provide a significant 40% improvement in data center energy efficiency within the existing facility and without the need to migrate software to a cloud architecture.
  • Generally, a 1-to-1 server upgrade (with no workload consolidation) to higher-capacity servers should be avoided. For the cost of deploying new servers, this example project only saw a 5% increase in data center energy efficiency.
  • Enabling server power management can cut data center energy use by 10%, which provides meaningful reductions in energy-related costs and carbon emissions even without the deployment of new hardware. When combined with a broader data center energy efficiency project, it can further boost energy performance.
  • When applied together with server energy efficiency improvements, power usage effectiveness (PUE) reductions can help accelerate data center energy efficiency.

Request an evaluation to view this report

Apply for a four-week evaluation of Uptime Intelligence; the leading source of research, insight and data-driven analysis focused on digital infrastructure.