DRL-based Slice Placement under Realistic Network Load Conditions - Archive ouverte HAL Access content directly
Conference Papers Year :

DRL-based Slice Placement under Realistic Network Load Conditions

Amina Boubendir
  • Function : Author
  • PersonId : 1080196
Fabice Guillemin
  • Function : Author
  • PersonId : 1080198
Pierre Sens

Abstract

We propose to demonstrate a network slice placement optimization solution based on Deep Reinforcement Learning (DRL), referred to as Heuristically-controlled DRL, which uses a heuristic to control the DRL algorithm convergence. The solution is adapted to realistic networks with large scale and under non-stationary traffic conditions (namely, the network load). We demonstrate the applicability of the proposed solution and its higher and stable performance over a non-controlled DRL-based solution. Demonstration scenarios include full online learning with multiple volatile network slice placement request arrivals.
Fichier principal
Vignette du fichier
Demo_CNSM_2021.pdf (810.17 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03516310 , version 1 (07-01-2022)

Identifiers

  • HAL Id : hal-03516310 , version 1

Cite

Jose Jurandir Alves Esteves, Amina Boubendir, Fabice Guillemin, Pierre Sens. DRL-based Slice Placement under Realistic Network Load Conditions. CNSM 2021 - 17th International Conference on Network and Service Management, Oct 2021, Izmir, Turkey. ⟨hal-03516310⟩
37 View
55 Download

Share

Gmail Facebook Twitter LinkedIn More