Energy proportionality is a measure of the relationship between power consumed in a computer system, and the rate at which useful work is done (its utilization). If the overall power consumption is proportional to the computer’s utilization, then the machine is said to be energy proportional.

Wikipedia: Energy proportional computing

Ideally, systems should consume power proportional to their utilization — using minimal energy at low load and scaling up smoothly under higher load.

The concept was introduced by Google engineers Luiz André Barroso and Urs Hölzle:

Energy-proportional designs would enable large energy savings in servers, potentially doubling their efficiency in real-life use. Achieving energy proportionality will require significant improvements in the energy usage profile of every system component, particularly the memory and disk subsystems.

Barroso, L.A. and Hölzle, U. (2007). The Case for Energy-Proportional Computing. Computer, vol. 40, no. 12, pp. 33-37.

Why it matters

Traditional hardware often exhibits poor energy proportionality: a server at 10% utilization may already consume 50% of its peak power draw due to static power consumption. Since servers in modern data centers typically operate at 10-50% utilization, this inefficiency has significant environmental and cost impact.

Architectural relevance

Virtualization, containerization, and cloud-native architectures can improve energy proportionality by consolidating workloads onto fewer, more highly utilized servers. However, achieving good energy proportionality requires deliberate architectural attention — it is a distinct quality concern beyond general energy efficiency.

The operating system’s role in enabling energy proportionality is a major area of investigation. Energy-aware scheduling algorithms consolidate workloads, allow CPU cores to enter deep sleep states, and coordinate power management across different system components.

Wikipedia: Energy proportional computing

Further reading