TENORAN: Automating Fine-grained Energy Efficiency Profiling in Open RAN Systems

TL;DR

TENORAN addresses the challenge of measuring energy consumption in disaggregated Open RAN by providing an automated end-to-end profiling framework that integrates container-level, server-level, and RU-level measurements, enabling synchronized power and performance data collection. The framework orchestrates deployment and execution from high-level test specifications on OpenShift, leveraging Tekton and ArgoCD, and collects per-component energy metrics using Kepler, Raritan PX4, and Yocto-Watt, with data stored centrally for analysis. Experimental results reveal stack-dependent energy behavior (OpenAirInterface vs srsRAN), linear core-network power with load, and meaningful end-to-end energy efficiency variations when xApps are in the loop, demonstrating the value of fine-grained profiling for energy-aware optimization. Overall, TENORAN provides a practical, automated foundation for data-driven energy–performance trade-offs in O-RAN deployments and paves the way for energy-aware future 5G/6G systems, including detailed per-component analysis and repeatable experiments with heterogeneous hardware.

Abstract

The transition to disaggregated and interoperable Open Radio Access Network (RAN) architectures and the introduction of RAN Intelligent Controllers (RICs) in O-RAN creates new resource optimization opportunities and fine-grained tuning and configuration of network components to save energy while fulfilling service demand. However, unlocking this potential requires fine-grained and accurate energy measurements across heterogeneous deployments. Three factors make this particularly challenging [...]. To address these challenges, we design the TENORAN framework, an automated measurement scaffold for fine-grained energy efficiency profiling of O-RAN deployments, and prototype it on a heterogeneous OpenShift cluster. TENORAN instruments an end-to-end deployment based on high-level specifications (e.g., gNB software stack and split options, traffic profiles), and collects synchronized performance metrics and power measurements for individual RAN components while the network is under controlled workloads including over-the-air traffic. Our experimental results demonstrate energy profiling of end-to-end experiments with xApps in the loop, energy efficiency differences between two RAN stacks, OpenAirInterface and srsRAN, in uplink and downlink, and core network power consumption trends.

Figures (7)

• Figure 1: High-level overview of TENORAN pipeline to automatically collect performance and power measurements.
• Figure 2: Foxconn RU power as measured by the Yocto-Watt.
• Figure 3: gNB power measurements under different traffic loads.
• Figure 4: Energy efficiency for and srsRAN.
• Figure 5: Power consumption of the core network UPF pod under different UDP loads.