Design Of a Reinforcement Learning-Driven Scheduler For Efficient And Frugal Container Orchestr[...]

CEA; Palaiseau; Sur place; EUR 40 000 - 60 000; Plein temps

Résumé du poste

A leading research organization in Saclay is offering an internship for a motivated student in Computer Science or related fields. The goal is to design and evaluate a reinforcement learning-driven scheduling strategy for container orchestration. Candidates should have a strong interest in machine learning and distributed systems, alongside problem-solving abilities. This position offers valuable experience in cutting-edge technology.

Qualifications

• Final year student in Computer Science, Artificial Intelligence, or related field.
• Strong interest in machine learning and distributed systems.
• Curiosity and creativity in problem-solving.

Responsabilités

• Explore the orchestration framework developed within the team.
• Conduct a state-of-the-art study on RL-based scheduling.
• Design, implement, and train a new RL-based scheduler.
• Develop a feature extraction module for container behavior.
• Evaluate the approach through experiments and benchmarks.

Connaissances

Interest in machine learning

Understanding of distributed systems

Problem-solving abilities

Creativity

Formation

Master’s or Engineering program in Computer Science or related field

Description du poste

Design of a Reinforcement Learning–Driven Scheduler for Efficient and Frugal Container Orchestration H/F

Category

Engineering science

Contract

Internship

Job location

Saclay

Subject

Context: Modern distributed systems (such as cloud and edge computing platforms) rely on orchestration frameworks like Kubernetes or Docker Swarm to manage the deployment and execution of applications. A key challenge in these environments is how to schedule containers efficiently, deciding which node should run each task, while balancing performance, energy efficiency, and resource usage.

6 months

Objective

The goal of this internship is to design and evaluate a new intelligent scheduling strategy using reinforcement learning (RL). The idea is to enable the system to learn how to make smarter scheduling decisions over time, optimizing container placement and sizing, dynamic resource allocation, response time and energy consumption, and even inter-container dependencies such as shared data or communication patterns.

Your missions

• Explore and understand the orchestration framework developed within the team.
• Conduct a state-of-the-art study on RL-based scheduling in cloud and distributed environments.
• Design, implement, and train a new RL-based scheduler.
• Develop a feature extraction module to characterize container behavior and guide the RL agent’s decisions.
• Evaluate your approach through experiments and benchmark comparisons.

Profile sought

We are looking for a motivated student in the final year of a Master’s or Engineering program in Computer Science, Artificial Intelligence, or a related field, with Interest in machine learning and distributed systems, Curiosity, creativity, and strong problem-solving abilities.