Aug 2026· Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication· 0 citations· 7 references
Computer Science
TL;DR
Preliminary numerical simulations with synthetic multi-slice workloads suggest that predictive orchestration can mitigate delayed adaptation, at the cost of controlled overprovisioning.
Abstract
Network slicing enables heterogeneous services to share a common 5G infrastructure, but reactive slice elasticity often scales resources only after service degradation has occurred. We propose predictive hierarchical orchestration for multi-domain network slicing, combining short term workload prediction, SLA feedback, and coordinated scaling across RAN, core, and transport domains. Preliminary numerical simulations with synthetic multi-slice workloads show that, in a mixed scenario, the proposed controller reduces the SLA violation rate from 56.7% to 5.0% compared with reactive-only scaling, while reducing reaction time from 4.86 to 1.50 control intervals. These initial results suggest that predictive orchestration can mitigate delayed adaptation, at the cost of controlled overprovisioning.
METIS is a declarative slice orchestrator that manages the Day-0/1/2 lifecycle of network slice instances through cascaded reconciliation loops, and derives 3GPP-aligned slice profiles via hierarchical aggregation following the 5G quality-of-service model.
Arman Divband, A. Yaghoubian, Navid Nikaein· arXiv.org· 0 citations
Real-time inter-slice resource allocation in the Radio Access Network (RAN) is a critical control function in 5G and emerging 6G networks, where the scheduler in the Distributed Unit (DU) dynamically allocates physical resources, namely Physical Resource Blocks (PRBs), to different network slices to meet their diverse...
Asim Zoulkarni, C. Papagianni, Georgios Iosifidis et al.· International Mediterranean...· 0 citations
A QoE-aware framework for Multi-Access Edge Computing-enabled Open Radio Access Network (O-RAN) architectures, combining a graph attention network (GAT) encoder, distributed multi-agent DRL, and privacy-preserving FL, while transitioning control from Quality of Service (QoS) to QoE metrics is proposed.
Manoj Prasad Kunasegran, Wai Leong Pang, S. K. Phang· IEEE Access· 0 citations
SAGF (SLO-Aware Graph Forecasting), an integrated prediction-and-control framework for proactive autoscaling, encodes six per-service metrics on a spatio-temporal call graph, predicts per-service SLO risk via GCN with temporal gating, and feeds these scores to a constrained RL controller for dependency-aware scaling.
Kang Yang, Feng Wang, Ruili Jiang· 2026 8th International Confe...· 2 citations
As 5G and Beyond networks increasingly expose programmable capabilities to vertical industries through standardized APIs such as CAMARA, new challenges emerge regarding efficient and reliable resource orchestration. In this paper we present a vertical-aware orchestration framework based on intelligent edge-deployed net...
Vincent Charpentier, Andreas Gavrielides, Miguel Camelo et al.· International Conference on...· 0 citations
Compute and memory resources in cloud environments are strictly managed and isolated by the control plane; in contrast, network resources lack equivalent management and isolation mechanisms. This best-effort treatment of networking leads to significant challenges for modern AI workloads, which have diverse and bandwidt...
