QoS-Driven Auto-Scaling Docker Containers Testbed with Dynamic Weighted Load Balancing

In the last decade, containerization technologies have ushered in a new era in the development, deployment, and management of applications. They have become the foundation for creating flexible, efficient, and scalable architectures, enabling faster deployment and resource utilization optimization. However, as applications grow in size and complexity, challenges arise in effectively managing computational resources, ensuring quality of service (QoS), and dynamically scaling in response to changes in workload. In this context, our work proposes the development of a testbed with auto-scaling docker containers that not only takes into account load balancing but also actively responds to changes in QoS. The key idea is to extend the concept of container auto-scaling towards self-adaptation to the system's and users' needs. We propose a load balancing technique that utilizes dynamic weight adjustments, automatically adapting to changes in workload and the number of deployed containers. This approach ensures an appropriate level of service through optimal container deployment. In the future, we will aim to develop a machine learning-enabled framework for deploying efficient, flexible, and cost-effective computing infrastructure that will meet the needs of different users and contribute to the sustainable development of cloud service providers.