A curated list of Deep Reinforcement Learning (DRL) research and related works on:
- Resource scheduling / placement in cloud environments
- Service autoscaling (microservice scaling-in / scaling-out)
- Cloud-level scheduling decisions involving task placement and resource allocation
With the increasing complexity of cloud environments and the dynamic nature of workloads, traditional scheduling algorithms often fall short in optimizing resource allocation. Deep Reinforcement Learning (DRL) offers a promising approach by learning optimal scheduling policies through environment interaction.
This repository collects significant works that leverage DRL for cloud scheduling and autoscaling, as well as related research in machine learning, heuristics, workload analysis, and system design.
The complete paper catalog (currently ~219 entries) is maintained in PAPER-LIST.md:
-
PAPER-LIST.md — full list of papers with authors, publication venue, summaries, and code availability, organized into:
- DRL — Resource Scheduling
- DRL — Autoscaling
- DRL — Edge Computing
- DRL — Robustness & Security
- Beyond DRL — Benchmarks & Simulators, Heuristic & Evolutionary Scheduling, ML Predictive Autoscaling, Microservice Architecture & Analysis, Other, Surveys & Reviews, System Design & Platform, Traditional Autoscaling, Traditional Resource Management, Workload Analysis & Characterization
To add or update a paper entry, edit PAPER-LIST.md directly.
Open-source code and artifacts related to the listed papers:
- CODE-LIST.md — curated links to open-source code repositories accompanying the papers
Code/— local folder for storing collected open-source code files
This repository is inspired by the awesome list movement. For broader DRL resources, see awesome-deep-reinforcement-learning.