Context Aware Task Orchestration With Deep Reinforcement Learning in Real Time Fog Computing Simulation Environment
In the ever-evolving landscape of cloud computing, fog and edge computing have become more prominent because of their natural property of proximity to demanding parts. With all the heterogeneity on the task processor side, generators have a variety of requirements in terms of complexity and latency....
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11003872/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In the ever-evolving landscape of cloud computing, fog and edge computing have become more prominent because of their natural property of proximity to demanding parts. With all the heterogeneity on the task processor side, generators have a variety of requirements in terms of complexity and latency. This study gathered all these different dimensions together in a robust ecosystem that can simulate a large number of various scenarios that meet changing requirements for task coordination and serves as a versatile platform for exploring strategies. A comprehensive multi-layered cloud simulation model that intricately considers both low-level edge/fog and cloud environment constraints is proposed. At the core of this contribution lies a novel task orchestration model that transforms the orchestration process into reinforcement learning training steps. The proposed approach offers substantial advantages in terms of task succession, energy efficiency, and resource utilization. The system was evaluated in a simulation developed under different fog computing environmental conditions in terms of edge device and task density. The results of the experiment proved the superiority of the proposed system over the existing round-robin heuristic-based algorithms and the offloading of policy baselines with an overall increase in precision 28%. In addition, a smart action space reduction technique is introduced to reduce the complexity of the action space, the proposed technique simplifies the decision-making process, leading to faster convergence and improved training efficiency. |
|---|---|
| ISSN: | 2169-3536 |