Orchestrating Energy-Efficient vRANs: Bayesian Learning and Experimental Results

Abstract

Virtualized base stations (vBS) can be implemented in diverse commodity platforms and are expected to bring unprecedented operational flexibility and cost efficiency to the next generation of cellular networks. However, their widespread adoption is hampered by their complex configuration options that affect in a non-traditional fashion both their performance and their power consumption. Following an in-depth experimental analysis in a bespoke testbed, we characterize the vBS power consumption profile and reveal previously unknown couplings between their various control knobs. Motivated by these findings, we develop a Bayesian learning framework for the orchestration of vBSs and design two novel algorithms: (i) BP-vRAN, which employs online learning to balance the vBS performance and energy consumption, and (ii) SBP-vRAN, which augments our optimization approach with safe controls that maximize performance while respecting hard power constraints. We show that our approaches are data-efficient, i.e., converge an order of magnitude faster than state-of-the-art Deep Reinforcement Learning methods, and achieve optimal performance. We demonstrate the efficacy of these solutions in an experimental prototype using real traffic traces.

Network Intelligence for Virtualized RAN Orchestration: The DAEMON Approach

“…We next present two algorithms developed within the context of the DAEMON project and fully detailed in [11] and [14], respectively, which handle vRAN orchestration and control operations at different timescales. We will use these algorithms as an example of the approaches that may be deployed concurrently, which may benefit from the framework proposed in this paper.…”

“…Our goal is to use O-RAN's control architecture to implement near-real-time configuration policies that are adaptive to system dynamics while satisfying hard energy constraints. Specifically, we consider the Safe Bayesian Optimization vRAN control algorithm (SBP-vRAN) recently introduced in [12], [14].…”

“…Other parametric methods, such as Reinforcement Learning relying on neural networks, need to be re-trained if the constraint changes, which substantially increases the required training data. The details of the learning model can be found in [12], [14].…”

Constrained Deep Reinforcement Based Functional Split Optimization in Virtualized RANs

“…The authors in [9] have proposed a learning framework that successfully solves a contextual bandit problem of dynamic computing and radio resource controls in vRANs using a deep reinforcement learning (RL) paradigm. Further, they leveraged Bayesian learning for energy efficient-based resource orchestrator in [10]. MLbased predictor also has been developed in [11] that learns to share the CPU resources between a vRAN workflow and other workflows in the same server.…”

“…On the other hand, optimizing the functional splits produces a high degree of complexity. In addition to the mentioned challenges, unlike legacy RANs, the behaviour of vRAN system performance such as computing utilization [9] and power consumption [10] is highly non-trivial. This non-triviality is also reinforced by vRAN deployment over the same platform with other workloads such as video analytics [11].…”