Hi, I’m Menra Romial

I’m a PhD student (also published as Romial Menra) at Inria Rennes and IMT Atlantique (Nantes), working on power management in cloud infrastructure, energy-aware computing systems, and sustainable distributed systems. My research focuses on Kubernetes orchestration, efficient server energy management, and optimizing energy consumption based on actual workload demands.

Data centers consume enormous amounts of energy globally, yet significant waste persists due to inefficient resource utilization and suboptimal orchestration strategies. I am driven by the challenge of designing systems that are both performant and energy-efficient - bridging the gap between server capabilities and software-level decision-making.

Kubernetes · Server Energy Management · Workload-Aware Optimization · Cloud Infrastructure · Power Management

If you want to discuss cloud computing, energy-aware systems, Kubernetes, or sustainable computing, I am always happy to connect: itsme [at] menraromial [dot] com

News

[All news →]

Selected publications

Euro-Par 2026 🏆 Best Paper Award nominee

Understanding Power Limiting Mechanisms in Modern Processors: A Deep Dive into Intel RAPL and Turbo Boost Dynamics

Romial Menra, Guillaume Rosinosky, Remous-Aris Koutsiamanis, Sébastien Bolle, Jean-Marc Menaud
Euro-Par 2026 - European Conference on Parallel and Distributed Computing · Acceptance rate: 26.2%
Abstract

Hardware-based power capping is essential for managing electrical consumption constraints and operational costs in data centers and High-Performance Computing. While the Intel Running Average Power Limit (RAPL) interface offers fine-grained control capabilities on modern servers, the interaction between its configuration parameters and the hardware’s dynamic performance boost mechanisms remains complex and often misunderstood. This paper provides a comprehensive understanding of these mechanisms and their effects on power consumption and Quality of Service. First, we conduct a sensitivity analysis to characterize how each parameter influences the processor’s throttling behavior. Based on the underlying exponential weighted moving average logic, we derive an analytical model to predict the effective duration of high-performance states. We validate this model through extensive experiments on four Intel Xeon microarchitectures. Furthermore, we quantify the impact of power capping on overall energy efficiency. Our experiments confirm that restricting power does not systematically lead to energy savings, highlighting a critical efficiency threshold below 50% of the Thermal Design Power, where the increase in execution time severely outweighs the power reduction. This work provides system administrators and researchers with the necessary insights to effectively configure power constraints while avoiding performance pitfalls.

COMPAS 2024

Intégration de l'Aspect Énergétique dans Kubernetes

Romial Menra, Remous-Aris Koutsiamanis, Jean-Marc Menaud
COMPAS 2024 - Conférence francophone d'informatique en Parallélisme, Architecture et Système, Nantes, France
Abstract

Les centres de données consomment une part croissante de l’énergie mondiale, souvent de manière inefficace en raison d’une mauvaise adéquation entre les ressources allouées et la consommation réelle. Nous proposons d’intégrer la gestion énergétique directement dans Kubernetes en exploitant les interfaces RAPL (Running Average Power Limit) des processeurs modernes. Notre approche permet un ordonnancement des conteneurs sensible à la consommation énergétique réelle du serveur, réduisant le gaspillage sans compromettre les performances applicatives.

[All publications →]

Latest posts

[All posts →]