OREO: O-RAN intElligence Orchestration of xApp-based network services

TL;DR

OREO introduces an NFV-inspired O-RAN xApp orchestrator that treats services as compositions of shareable, semantically equivalent functions implemented by xApps at varying complexity. By formulating the xDeSh problem and solving it with a Lagrangian-relaxation based heuristic, OREO balances service quality and latency targets against a constrained resource budget, enabling sharing of xApps across services to maximize served demand. Numerical analyses show OREO remains close to the optimum (within about 0.75–0.86 of the best possible) while using substantially fewer xApps and less CPU resources than baselines; a proof-of-concept testbed demonstrates a 37.5% CPU reduction versus the state of the art. Overall, OREO advances practical, scalable O-RAN orchestration by leveraging NFV-based function decomposition, targeted complexity levels, and resource-aware sharing to improve service density and efficiency in real-world deployments.

Abstract

The Open Radio Access Network (O-RAN) architecture aims to support a plethora of network services, such as beam management and network slicing, through the use of third-party applications called xApps. To efficiently provide network services at the radio interface, it is thus essential that the deployment of the xApps is carefully orchestrated. In this paper, we introduce OREO, an O-RAN xApp orchestrator, designed to maximize the offered services. OREO's key idea is that services can share xApps whenever they correspond to semantically equivalent functions, and the xApp output is of sufficient quality to fulfill the service requirements. By leveraging a multi-layer graph model that captures all the system components, from services to xApps, OREO implements an algorithmic solution that selects the best service configuration, maximizes the number of shared xApps, and efficiently and dynamically allocates resources to them. Numerical results as well as experimental tests performed using our proof-of-concept implementation, demonstrate that OREO closely matches the optimum, obtained by solving an NP-hard problem. Further, it outperforms the state of the art, deploying up to 35% more services with an average of 30% fewer xApps and a similar reduction in the resource consumption.

Figures (7)

• Figure 1: OREO design and integration in the O-RAN architecture. The workflow (dashed black line) is as follows: ($i$) the MNO submits service requests via the Human-Machine interface (HMI); ($ii$) OREO processes such requests and, with the support of the xApp lifecycle manager, instructs the near-RT RIC via the O1 interface about which xApps to deploy.
• Figure 2: Graph-based representation of the system under study and relation between its main components. For clarity, the xApp layer only includes the xApps implementing function $f_1$.
• Figure 3: The xDeSh problem is solved with an iterative algorithm that alternates the Lagrangian relaxation and the subgradient method until the set stopping criterion is met.
• Figure 4: Numerical results: Percentage of services (top) and xApps (bottom) deployed by Optimal, OREO, and OrchestRAN.
• Figure 5: Numerical results: CPU (left), RAM (center), and Disk (right) resources used by Optimal, OREO, and OrchestRAN.

Theorems & Definitions (4)

• Proposition 1
• proof
• Proposition 2
• proof