PadalaSravan14/drl_edge_cloud_scheduler: v1.0 – DRL-Based Multi-Objective Task Scheduling for Edge-Cloud Computing
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This release contains the official implementation of the research work: "DRL-Based Multi-Objective Task Scheduling for Edge-Cloud Computing: Latency, Energy, and SLA Optimisation". The repository implements an Adaptive-Weight Deep Q-Network (AW-DQN) framework for intelligent task scheduling across edge and cloud environments. The system models scheduling as a Markov Decision Process (MDP) and jointly optimises multiple objectives including latency minimisation, energy efficiency, and SLA compliance. Key features of this release include: Full DRL training pipeline (DQN, DDQN, PPO) Adaptive multi-objective reward mechanism State representation with task, resource, and network features Support for synthetic workloads (Poisson-based) and real-world datasets Integration with Google Cluster Traces and Azure Functions datasets Baseline implementations (FIFO, Round-Robin, Min-Min, Greedy-Energy) Evaluation framework with statistical validation (t-tests, confidence intervals) Experiment scripts for scalability, ablation, and fault tolerance analysis This release ensures full reproducibility of the experimental results reported in the manuscript and provides a modular, extensible framework for future research in edge-cloud scheduling using Deep Reinforcement Learning.
Fields of Study & Subjects
IoT and Edge/Fog Computing
Cloud Computing and Resource Management
Energy Efficiency in Computing
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 30 March 2026
- Language
- EN