Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Graph neural network for in-network placement of real-time metaverse tasks in next-generation network

Sulaiman Muhammad Rashid, Ibrahim Aliyu, Il-Kwon Jeong, Tai-Won Um, Jinsul Kim

TL;DR

The paper tackles real-time metaverse rendering under delay constraints by proposing an SDN-enabled in-network computing (INC) framework and a graph neural network (GNN) that learns near-optimal task placement. The ILP-based optimizer generates offline labels to train the GNN, which operates on a graph of tasks and resources using both neighbor-aware and independent convolutions to handle connected and isolated nodes. Empirical results show the GNN achieving an accuracy of $0.9244$ (92.44%) and far faster inference than the exact ILP solution (about $100\times$ faster), with analysis of task splitting vs no-splitting and MEC/COIN offloading under varying delay constraints. The approach offers a scalable, near-optimal solution for delay-constrained metaverse rendering and can be extended to other real-time edge computing scenarios.

Abstract

This study addresses the challenge of real-time metaverse applications by proposing an in-network placement and task-offloading solution for delay-constrained computing tasks in next-generation networks. The metaverse, envisioned as a parallel virtual world, requires seamless real-time experiences across diverse applications. The study introduces a software-defined networking (SDN)-based architecture and employs graph neural network (GNN) techniques for intelligent and adaptive task allocation in in-network computing (INC). Considering time constraints and computing capabilities, the proposed model optimally decides whether to offload rendering tasks to INC nodes or edge server. Extensive experiments demonstrate the superior performance of the proposed GNN model, achieving 97% accuracy compared to 72% for multilayer perceptron (MLP) and 70% for decision trees (DTs). The study fills the research gap in in-network placement for real-time metaverse applications, offering insights into efficient rendering task handling.

Graph neural network for in-network placement of real-time metaverse tasks in next-generation network

TL;DR

The paper tackles real-time metaverse rendering under delay constraints by proposing an SDN-enabled in-network computing (INC) framework and a graph neural network (GNN) that learns near-optimal task placement. The ILP-based optimizer generates offline labels to train the GNN, which operates on a graph of tasks and resources using both neighbor-aware and independent convolutions to handle connected and isolated nodes. Empirical results show the GNN achieving an accuracy of (92.44%) and far faster inference than the exact ILP solution (about faster), with analysis of task splitting vs no-splitting and MEC/COIN offloading under varying delay constraints. The approach offers a scalable, near-optimal solution for delay-constrained metaverse rendering and can be extended to other real-time edge computing scenarios.

Abstract

This study addresses the challenge of real-time metaverse applications by proposing an in-network placement and task-offloading solution for delay-constrained computing tasks in next-generation networks. The metaverse, envisioned as a parallel virtual world, requires seamless real-time experiences across diverse applications. The study introduces a software-defined networking (SDN)-based architecture and employs graph neural network (GNN) techniques for intelligent and adaptive task allocation in in-network computing (INC). Considering time constraints and computing capabilities, the proposed model optimally decides whether to offload rendering tasks to INC nodes or edge server. Extensive experiments demonstrate the superior performance of the proposed GNN model, achieving 97% accuracy compared to 72% for multilayer perceptron (MLP) and 70% for decision trees (DTs). The study fills the research gap in in-network placement for real-time metaverse applications, offering insights into efficient rendering task handling.
Paper Structure (26 sections, 20 equations, 8 figures, 4 tables)

This paper contains 26 sections, 20 equations, 8 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Motivation and Contributions
  3. Related work
  4. In-network placement of delay-constraint tasks
  5. ML for SDN-based task placement
  6. System Model
  7. Task Model
  8. Computing Model
  9. Optimization Problem Formulation
  10. Optimal task placement using GNN
  11. Graph Construction
  12. Feature Embedding
  13. Graph Convolution
  14. Neighbor-Aware Convolution
  15. Independent Convolution
  16. ...and 11 more sections

Figures (8)

  • Figure 1: COIN-MEC metaverse task offloading scenario
  • Figure 2: Sequence diagram of rendering request and INC procedures
  • Figure 3: GNN-based Adaptive Decision Network
  • Figure 4: Time complexity analysis of Algorithm computation of the ML models as compared to the optimal solution
  • Figure 5: Performance Gain vs Rate of Request
  • ...and 3 more figures