DRL-based Slice Placement under Realistic Network Load Conditions
This work addresses network slice placement for telecommunications under dynamic load conditions, representing an incremental improvement over existing DRL methods.
The authors tackled the problem of optimizing network slice placement under realistic, non-stationary traffic conditions by proposing a Heuristically-controlled Deep Reinforcement Learning (DRL) solution, which demonstrated higher and more stable performance compared to a non-controlled DRL-based approach in scenarios with volatile request arrivals.
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.