arXiv subjects

Sophie Cerf

Publications and source records attributed to Sophie Cerf.

6 recordsLinked to original sources

arXiv2026

LPV Control for Dynamic Power Capping in High-Performance Computing under Mixed Workloads

Balancing energy consumption and performance remains a critical challenge in High Performance Computing (HPC) systems. While static power capping mechanisms such as Intel's Running Average Power Limit (RAPL) offer basic control capabilities, they lack the flexibility to adapt to dynamically varying workloads. This work studies dynamic power regulation for mixed workload scenarios. We investigate two feedback strategies: a gain scheduled proportional-integral (PI) controller and a polytopic linear parameter-varying (LPV) controller synthesized via H$\infty$ control, both scheduled by a workload indicator that changes between memory and compute phase. We evaluate tracking performance, phase switching, and robustness under practical power cap constraints. While both controllers respect power limits, the LPV design achieves lower tracking error, lower control variance, and smoother transients during phase changes than gain scheduled PI.

cs.DC↗

arXiv2026

Fancy Some Chips for Your TeaStore? Modeling the Control of an Adaptable Discrete System

When designing new web applications, developers must cope with different kinds of constraints relative to the resources they rely on: software, hardware, network, online micro-services, or any combination of the mentioned entities. Together, these entities form a complex system of communicating interdependent processes, physical or logical. It is very desirable that such system ensures its robustness to provide a good quality of service. In this paper we introduce Chips, a language that aims at facilitating the design of models made of various entwined components. It allows the description of applications in the form of functional blocks. Chips mixes notions from control theory and general purpose programming languages to generate robust component-based models. This paper presents how to use Chips to systematically design, model and analyse a complex system project, using a variation of the Adaptable TeaStore application as running example.

cs.PL↗

arXiv2025

Mitigating Shared Storage Congestion Using Control Theory

Efficient data access in High-Performance Computing (HPC) systems is essential to the performance of intensive computing tasks. Traditional optimizations of the I/O stack aim to improve peak performance but are often workload specific and require deep expertise, making them difficult to generalize or re-use. In shared HPC environments, resource congestion can lead to unpredictable performance, causing slowdowns and timeouts. To address these challenges, we propose a self-adaptive approach based on Control Theory to dynamically regulate client-side I/O rates. Our approach leverages a small set of runtime system load metrics to reduce congestion and enhance performance stability. We implement a controller in a multi-node cluster and evaluate it on a real testbed under a representative workload. Experimental results demonstrate that our method effectively mitigates I/O congestion, reducing total runtime by up to 20% and lowering tail latency, while maintaining stable performance.

cs.DC↗

arXiv2021

Sustaining Performance While Reducing Energy Consumption: A Control Theory Approach

Production high-performance computing systems continue to grow in complexity and size. As applications struggle to make use of increasingly heterogeneous compute nodes, maintaining high efficiency (performance per watt) for the whole platform becomes a challenge. Alongside the growing complexity of scientific workloads, this extreme heterogeneity is also an opportunity: as applications dynamically undergo variations in workload, due to phases or data/compute movement between devices, one can dynamically adjust power across compute elements to save energy without impacting performance. With an aim toward an autonomous and dynamic power management strategy for current and future HPC architectures, this paper explores the use of control theory for the design of a dynamic power regulation method. Structured as a feedback loop, our approach-which is novel in computing resource management-consists of periodically monitoring application progress and choosing at runtime a suitable power cap for processors. Thanks to a preliminary offline identification process, we derive a model of the dynamics of the system and a proportional-integral (PI) controller. We evaluate our approach on top of an existing resource management framework, the Argo Node Resource Manager, deployed on several clusters of Grid'5000, using a standard memory-bound HPC benchmark.

cs.DC↗

arXiv2020

Event-Based Control for Online Training of Neural Networks

Convolutional Neural Network (CNN) has become the most used method for image classification tasks. During its training the learning rate and the gradient are two key factors to tune for influencing the convergence speed of the model. Usual learning rate strategies are time-based i.e. monotonous decay over time. Recent state-of-the-art techniques focus on adaptive gradient algorithms i.e. Adam and its versions. In this paper we consider an online learning scenario and we propose two Event-Based control loops to adjust the learning rate of a classical algorithm E (Exponential)/PD (Proportional Derivative)-Control. The first Event-Based control loop will be implemented to prevent sudden drop of the learning rate when the model is approaching the optimum. The second Event-Based control loop will decide, based on the learning speed, when to switch to the next data batch. Experimental evaluationis provided using two state-of-the-art machine learning image datasets (CIFAR-10 and CIFAR-100). Results show the Event-Based E/PD is better than the original algorithm (higher final accuracy, lower final loss value), and the Double-Event-BasedE/PD can accelerate the training process, save up to 67% training time compared to state-of-the-art algorithms and even result in better performance.

cs.LG↗

arXiv2019

Feedback Control for Online Training of Neural Networks

Convolutional neural networks (CNNs) are commonly used for image classification tasks, raising the challenge of their application on data flows. During their training, adaptation is often performed by tuning the learning rate. Usual learning rate strategies are time-based i.e. monotonously decreasing. In this paper, we advocate switching to a performance-based adaptation, in order to improve the learning efficiency. We present E (Exponential)/PD (Proportional Derivative)-Control, a conditional learning rate strategy that combines a feedback PD controller based on the CNN loss function, with an exponential control signal to smartly boost the learning and adapt the PD parameters. Stability proof is provided as well as an experimental evaluation using two state of the art image datasets (CIFAR-10 and Fashion-MNIST). Results show better performances than the related works (faster network accuracy growth reaching higher levels) and robustness of the E/PD-Control regarding its parametrization.

cs.LG↗