<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Home on Menra Romial</title><link>https://menraromial.com/</link><description>Recent content in Home on Menra Romial</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Thu, 18 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://menraromial.com/index.xml" rel="self" type="application/rss+xml"/><item><title>Best Paper nomination at Euro-Par 2026</title><link>https://menraromial.com/news/best-paper-nomination-europar-2026/</link><pubDate>Thu, 18 Jun 2026 00:00:00 +0000</pubDate><guid>https://menraromial.com/news/best-paper-nomination-europar-2026/</guid><description>&lt;p&gt;Excited to share that our paper &amp;ldquo;&lt;em&gt;Understanding Power Limiting Mechanisms in Modern Processors: A Deep Dive into Intel RAPL and Turbo Boost Dynamics&lt;/em&gt;&amp;rdquo; has been nominated for the &lt;strong&gt;Best Paper Award&lt;/strong&gt; at &lt;strong&gt;Euro-Par 2026&lt;/strong&gt;!&lt;/p&gt;</description></item><item><title>Paper accepted at Euro-Par 2026</title><link>https://menraromial.com/news/europar-2026-paper-accepted/</link><pubDate>Thu, 30 Apr 2026 00:00:00 +0000</pubDate><guid>https://menraromial.com/news/europar-2026-paper-accepted/</guid><description>&lt;p&gt;Proud to announce that our paper &amp;ldquo;&lt;em&gt;Understanding Power Limiting Mechanisms in Modern Processors: A Deep Dive into Intel RAPL and Turbo Boost Dynamics&lt;/em&gt;,&amp;rdquo; co-authored with Guillaume Rosinosky, Remous-Aris Koutsiamanis, Sébastien Bolle, and Jean-Marc Menaud, has been accepted at &lt;strong&gt;Euro-Par 2026&lt;/strong&gt; (acceptance rate: 26.2%).&lt;/p&gt;</description></item><item><title>Understanding Power Limiting Mechanisms in Modern Processors: A Deep Dive into Intel RAPL and Turbo Boost Dynamics</title><link/><pubDate>Thu, 30 Apr 2026 00:00:00 +0000</pubDate><guid/><description>&lt;p&gt;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&amp;rsquo;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&amp;rsquo;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.&lt;/p&gt;</description></item><item><title>Welcoming Clement Obama as M2 Research Intern</title><link>https://menraromial.com/news/m2-intern-clement-obama/</link><pubDate>Thu, 16 Apr 2026 00:00:00 +0000</pubDate><guid>https://menraromial.com/news/m2-intern-clement-obama/</guid><description>&lt;p&gt;Welcoming &lt;strong&gt;Clement Obama&lt;/strong&gt; as a new M2 research intern (April–September 2026) working on &lt;em&gt;Reactive control of energy constraints in Kubernetes clusters: multi-lever re-adaptation policy with SLA management&lt;/em&gt;.&lt;/p&gt;</description></item><item><title>Volunteering at UCC &amp; BDCAT 2025 - My Experience</title><link>https://menraromial.com/news/volunteering-ucc-bdcat-2025/</link><pubDate>Thu, 04 Dec 2025 00:00:00 +0000</pubDate><guid>https://menraromial.com/news/volunteering-ucc-bdcat-2025/</guid><description>&lt;p&gt;From December 1st to 4th, I had the opportunity to volunteer at the 2025 editions of &lt;strong&gt;&lt;a href="https://ucc-conference.org/"&gt;UCC&lt;/a&gt;&lt;/strong&gt; (Utility and Cloud Computing) and &lt;strong&gt;&lt;a href="https://bdcat-conference.org/"&gt;BDCAT&lt;/a&gt;&lt;/strong&gt; (Big Data Computing, Applications and Technologies). Both conferences were hosted together this year, creating an inspiring environment filled with researchers, engineers, and students from around the world.&lt;/p&gt;</description></item><item><title>A Comprehensive Guide to powercap-utils on Linux</title><link>https://menraromial.com/posts/comprehensive-guide-powercap-utils-linux/</link><pubDate>Thu, 23 Oct 2025 13:00:00 +0000</pubDate><guid>https://menraromial.com/posts/comprehensive-guide-powercap-utils-linux/</guid><description>&lt;h3 id="1-introduction-what-is-powercap-utils"&gt;1. Introduction: What is &lt;code&gt;powercap-utils&lt;/code&gt;?&lt;/h3&gt;
&lt;p&gt;&lt;code&gt;powercap-utils&lt;/code&gt; is a suite of user-space command-line tools for Linux that allow you to interact with the kernel&amp;rsquo;s &lt;strong&gt;Power Capping Framework&lt;/strong&gt;. This framework provides a standardized way to monitor and limit the power consumption of hardware devices.&lt;/p&gt;
&lt;p&gt;Its most common use is to control &lt;strong&gt;Intel&amp;rsquo;s Running Average Power Limit (RAPL)&lt;/strong&gt; and &lt;strong&gt;AMD&amp;rsquo;s RAPL-like features&lt;/strong&gt; on modern CPUs. These features let you directly set power budgets for the CPU package, its cores, and DRAM.&lt;/p&gt;</description></item><item><title>Building a Kubernetes Controller with Kubebuilder from Scratch</title><link>https://menraromial.com/posts/building-a-kubernetes-controller-with-kubebuilder-from-scratch/</link><pubDate>Mon, 22 Sep 2025 10:30:00 +0000</pubDate><guid>https://menraromial.com/posts/building-a-kubernetes-controller-with-kubebuilder-from-scratch/</guid><description>&lt;p&gt;In this tutorial, you will learn how to build a custom Kubernetes controller (also known as an Operator) using Kubebuilder. We will create a Custom Resource Definition (CRD) for a simple &lt;code&gt;Website&lt;/code&gt; resource. The goal of our controller will be to watch for these &lt;code&gt;Website&lt;/code&gt; resources and, for each one, create a corresponding Kubernetes &lt;code&gt;Deployment&lt;/code&gt; and &lt;code&gt;Service&lt;/code&gt;.&lt;/p&gt;
&lt;p&gt;This is a powerful pattern that allows you to extend the Kubernetes API to manage your own applications declaratively.&lt;/p&gt;</description></item><item><title>Guide to Building a Custom Kubernetes Scheduler</title><link>https://menraromial.com/posts/guide-building-custom-kubernetes-scheduler/</link><pubDate>Sun, 24 Aug 2025 17:30:00 +0000</pubDate><guid>https://menraromial.com/posts/guide-building-custom-kubernetes-scheduler/</guid><description>&lt;p&gt;This guide will walk you through creating a custom scheduler plugin that influences where pods are placed based on custom logic. We will build a simple &amp;ldquo;Network-Aware&amp;rdquo; plugin that prefers to schedule pods on nodes with a specific network label.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;What We Will Build:&lt;/strong&gt;&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;A &lt;code&gt;Filter&lt;/code&gt; plugin:&lt;/strong&gt; It will only allow pods that request &lt;code&gt;high-speed&lt;/code&gt; networking to be scheduled on nodes that are labeled as &lt;code&gt;network-topology=high-speed&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;A &lt;code&gt;Score&lt;/code&gt; plugin:&lt;/strong&gt; It will give a higher score to nodes with the &lt;code&gt;high-speed&lt;/code&gt; label, making them more desirable.&lt;/li&gt;
&lt;li&gt;We will then package this plugin into a new scheduler binary, containerize it, and deploy it to a local Kubernetes cluster.&lt;/li&gt;
&lt;/ul&gt;
&lt;hr&gt;
&lt;h3 id="part-1-understanding-the-foundations---the-scheduling-framework"&gt;&lt;strong&gt;Part 1: Understanding the Foundations - The Scheduling Framework&lt;/strong&gt;&lt;/h3&gt;
&lt;p&gt;Before writing any code, it&amp;rsquo;s crucial to understand the &amp;ldquo;why&amp;rdquo; and &amp;ldquo;how&amp;rdquo;.&lt;/p&gt;</description></item><item><title>Master the CI/CD Pipeline Locally</title><link>https://menraromial.com/posts/master-ci-ci-pipeline-locally/</link><pubDate>Wed, 20 Aug 2025 10:30:00 +0000</pubDate><guid>https://menraromial.com/posts/master-ci-ci-pipeline-locally/</guid><description>&lt;p&gt;Welcome to this comprehensive course on DevOps culture and tooling. The goal is to guide you, step-by-step, from theory to practice, empowering you to build, secure, and automate a full Continuous Integration and Continuous Deployment (CI/CD) pipeline.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Course Philosophy:&lt;/strong&gt;
Everything we learn will be put into practice in a &lt;strong&gt;100% local lab environment&lt;/strong&gt;, running on your own machine using Docker. You will not need any cloud accounts or paid services. Every concept is illustrated with hands-on workshops and concrete examples.&lt;/p&gt;</description></item><item><title>Setting Up a Kubernetes Cluster with Kubeadm and CRI-O</title><link>https://menraromial.com/posts/kubernetes-with-kubeadm-and-cri.io/</link><pubDate>Tue, 25 Mar 2025 16:40:16 +0000</pubDate><guid>https://menraromial.com/posts/kubernetes-with-kubeadm-and-cri.io/</guid><description>&lt;h2 id="prerequisites"&gt;Prerequisites&lt;/h2&gt;
&lt;p&gt;Before setting up a Kubernetes cluster using Kubeadm and CRI-O, ensure that you meet the following requirements:&lt;/p&gt;
&lt;h3 id="hardware-requirements"&gt;Hardware Requirements&lt;/h3&gt;
&lt;ul&gt;
&lt;li&gt;At least &lt;strong&gt;2 GB of RAM&lt;/strong&gt; per node (4 GB recommended for the master node)&lt;/li&gt;
&lt;li&gt;At least &lt;strong&gt;2 CPUs&lt;/strong&gt; per node&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;20 GB of free disk space&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;A stable &lt;strong&gt;internet connection&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;All nodes should be able to communicate with each other&lt;/li&gt;
&lt;/ul&gt;
&lt;h3 id="software-requirements"&gt;Software Requirements&lt;/h3&gt;
&lt;ul&gt;
&lt;li&gt;A supported &lt;strong&gt;Linux distribution&lt;/strong&gt; (Ubuntu 20.04+ or Debian 10+ recommended)&lt;/li&gt;
&lt;li&gt;User access with &lt;strong&gt;sudo privileges&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;Basic understanding of &lt;strong&gt;Linux command-line interface (CLI)&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;Knowledge of &lt;strong&gt;networking concepts&lt;/strong&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;h2 id="step-1-configure-kernel-parameters"&gt;Step 1: Configure Kernel Parameters&lt;/h2&gt;
&lt;p&gt;Kubernetes requires some kernel parameters to be set to allow IPtables to see bridged traffic.&lt;/p&gt;</description></item><item><title>Setting Up a K3s Cluster with a Hello-World Deployment</title><link>https://menraromial.com/posts/setup-k3s-cluster/</link><pubDate>Sun, 07 Jul 2024 14:30:00 +0000</pubDate><guid>https://menraromial.com/posts/setup-k3s-cluster/</guid><description>&lt;p&gt;This tutorial will guide you through setting up a K3s cluster with one master node and two worker nodes, deploying a simple &amp;ldquo;Hello-World&amp;rdquo; application, and utilizing various &lt;code&gt;kubectl&lt;/code&gt; commands to manage and inspect Kubernetes resources.&lt;/p&gt;
&lt;h2 id="prerequisites"&gt;Prerequisites&lt;/h2&gt;
&lt;ul&gt;
&lt;li&gt;Three Linux machines (or VMs) with access to each other.&lt;/li&gt;
&lt;li&gt;Basic understanding of Linux commands and Kubernetes concepts.&lt;/li&gt;
&lt;/ul&gt;
&lt;h2 id="step-1-installing-k3s"&gt;Step 1: Installing K3s&lt;/h2&gt;
&lt;h3 id="install-k3s-on-the-master-node"&gt;Install K3s on the Master Node&lt;/h3&gt;
&lt;ol&gt;
&lt;li&gt;
&lt;p&gt;&lt;strong&gt;SSH into the Master Node:&lt;/strong&gt;&lt;/p&gt;
&lt;div class="highlight"&gt;&lt;pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"&gt;&lt;code class="language-bash" data-lang="bash"&gt;&lt;span style="display:flex;"&gt;&lt;span&gt;ssh user@master-node-ip
&lt;/span&gt;&lt;/span&gt;&lt;/code&gt;&lt;/pre&gt;&lt;/div&gt;&lt;/li&gt;
&lt;li&gt;
&lt;p&gt;&lt;strong&gt;Install K3s:&lt;/strong&gt;&lt;/p&gt;</description></item><item><title>Intégration de l'Aspect Énergétique dans Kubernetes</title><link/><pubDate>Tue, 02 Jul 2024 00:00:00 +0000</pubDate><guid/><description>&lt;p&gt;Les centres de données consomment une part croissante de l&amp;rsquo;énergie mondiale, souvent de manière inefficace en raison d&amp;rsquo;une mauvaise adéquation entre les ressources allouées et la consommation réelle. Nous proposons d&amp;rsquo;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.&lt;/p&gt;</description></item><item><title>CV</title><link>https://menraromial.com/cv/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://menraromial.com/cv/</guid><description>&lt;p&gt;&lt;strong&gt;Menra Wedwang Romial&lt;/strong&gt; - Cloud &amp;amp; Green IT - Application Deployment and Energy Optimization of Infrastructures&lt;/p&gt;
&lt;p&gt;itsme [at] menraromial [dot] com · &lt;a href="https://github.com/menraromial"&gt;GitHub&lt;/a&gt; · &lt;a href="https://www.linkedin.com/in/menraromial"&gt;LinkedIn&lt;/a&gt; · &lt;a href="https://orcid.org/0009-0007-0943-8593"&gt;ORCID&lt;/a&gt; · &lt;a href="https://menraromial.com/pdf/resume_romial.pdf"&gt;PDF version&lt;/a&gt;
Languages: French, English&lt;/p&gt;
&lt;p&gt;PhD student specializing in Cloud Computing and Green IT, with a strong focus on energy-efficient application deployment and infrastructure optimization. My expertise spans from low-level power management (Intel RAPL) to high-level orchestration systems (Kubernetes), with a proven track record in designing scalable, energy-aware solutions for modern cloud infrastructures.&lt;/p&gt;</description></item><item><title>Projects</title><link>https://menraromial.com/projects/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://menraromial.com/projects/</guid><description>&lt;h2 id="research"&gt;Research&lt;/h2&gt;
&lt;h3 id="thèse-20242027"&gt;Thèse (2024–2027)&lt;/h3&gt;
&lt;p&gt;Contribution à la gestion de la puissance dans les infrastructures cloud : de la limitation matérielle à l&amp;rsquo;orchestration adaptative. PhD research at Inria / IMT Atlantique.
&lt;code&gt;Research · Kubernetes · RAPL · Grid'5000 · Power Management · Cloud&lt;/code&gt; - &lt;a href="https://menraromial.com/publications/"&gt;publications&lt;/a&gt;&lt;/p&gt;
&lt;h3 id="powercap"&gt;Powercap&lt;/h3&gt;
&lt;p&gt;A power-capping tool leveraging Intel RAPL, designed to plug into Kubernetes as a DaemonSet. Enforces per-node power budgets at the package and DRAM domain level, with audit logs and dry-run mode for cautious operators.
&lt;code&gt;Go · Kubernetes · RAPL · Linux&lt;/code&gt;&lt;/p&gt;</description></item><item><title>Talks</title><link>https://menraromial.com/talks/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://menraromial.com/talks/</guid><description>&lt;h2 id="energy-aware-kubernetes-orchestration-for-grid-responsive-computing"&gt;Energy-Aware Kubernetes Orchestration for Grid-Responsive Computing&lt;/h2&gt;
&lt;p&gt;&lt;strong&gt;STACK Team Seminar&lt;/strong&gt; - Arzon, France · October 2025&lt;/p&gt;
&lt;p&gt;A comprehensive approach to adapting Kubernetes workloads to electricity grid fluctuations through constraint-based scheduling and dynamic scaling, addressing the challenges of electricity price volatility and grid stability in cloud computing environments.&lt;/p&gt;
&lt;p&gt;Related posts:&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;&lt;a href="https://menraromial.com/posts/guide-building-custom-kubernetes-scheduler/"&gt;Building a Custom Kubernetes Scheduler&lt;/a&gt;&lt;/li&gt;
&lt;li&gt;&lt;a href="https://menraromial.com/posts/building-a-kubernetes-controller-with-kubebuilder-from-scratch/"&gt;Building a Kubernetes Controller with Kubebuilder&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;
&lt;hr&gt;
&lt;h2 id="integration-of-the-energy-aspect-in-kubernetes"&gt;Integration of the Energy Aspect in Kubernetes&lt;/h2&gt;
&lt;p&gt;&lt;strong&gt;ComPAS 2024&lt;/strong&gt; - Faculty of Medicine, Nantes, France · July 2024&lt;/p&gt;</description></item><item><title>Teaching</title><link>https://menraromial.com/teaching/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://menraromial.com/teaching/</guid><description>&lt;h2 id="bases-de-données-databases"&gt;Bases de Données (Databases)&lt;/h2&gt;
&lt;p&gt;&lt;strong&gt;IMT Atlantique, Nantes&lt;/strong&gt; · Teaching Assistant (TD/TP) · Spring 2025&lt;/p&gt;
&lt;p&gt;TD/TP on relational databases for first-year engineering students (FISE 1A, semester S6):&lt;/p&gt;
&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Relational model&lt;/strong&gt; and data integrity constraints&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;SQL&lt;/strong&gt; queries, transactions, and security&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Normalization theory&lt;/strong&gt; (functional dependencies, normal forms)&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Conceptual modeling&lt;/strong&gt; with UML&lt;/li&gt;
&lt;li&gt;&lt;strong&gt;Programmatic interfaces&lt;/strong&gt; for data persistence in applications&lt;/li&gt;
&lt;/ul&gt;
&lt;hr&gt;
&lt;h2 id="cloud-computing-vmware-vsphere"&gt;Cloud Computing (VMware vSphere)&lt;/h2&gt;
&lt;p&gt;&lt;strong&gt;IMT Atlantique, Nantes&lt;/strong&gt; · Teaching Assistant (TP) · Spring 2025&lt;/p&gt;
&lt;p&gt;Hands-on lab sessions on cloud infrastructure using &lt;strong&gt;VMware vSphere&lt;/strong&gt; and &lt;strong&gt;vCenter&lt;/strong&gt;:&lt;/p&gt;</description></item></channel></rss>