arXiv subjects

Stéphan Plassart

Publications and source records attributed to Stéphan Plassart.

3 recordsLinked to original sources

arXiv2025

Falafels: A tool for Estimating Federated Learning Energy Consumption via Discrete Simulation

The growth in computational power and data hungriness of Machine Learning has led to an important shift of research efforts towards the distribution of ML models on multiple machines, leading in even more powerful models. However, there exists many Distributed Artificial Intelligence paradigms and for each of them the platform and algorithm configurations play an important role in terms of training time and energy consumption. Many mathematical models and frameworks can respectively predict and benchmark this energy consumption, nonetheless, the former lacks of realism and extensibility while the latter suffers high run-times and actual power consumption. In this article, we introduce Falafels, an extensible tool that predicts the energy consumption and training time of -but not limited to -Federated Learning systems. It distinguishes itself with its discrete-simulatorbased solution leading to nearly instant run-time and fast development of new algorithms. Furthermore, we show this approach permits the use of an evolutionary algorithm providing the ability to optimize the system configuration for a given machine learning workload.

cs.DC↗

arXiv2024

Saihu: A Common Interface of Worst-Case Delay Analysis Tools for Time-Sensitive Networks

Time-sensitive networks, as in the context of IEEE-TSN and IETF-Detnet, require bounds on worst-case delays. Various network analysis tools compute such bounds; these tools are based on different methods and provide delay bounds that are all valid but may differ; furthermore, it is generally not known which tool will provide the best bound. To obtain the best possible bound, users need to implement multiple pieces of code with a different syntax for every tool, which is impractical and error-prone. To address this issue, we present Saihu, a Python interface that integrates the three most frequently used worst-case network analysis tools: xTFA, DiscoDNC, and Panco. They altogether implement six analysis methods. Saihu provides a general interface that enables defining a network in a single file and executing all tools simultaneously without any modification. Saihu further exports analysis results as formatted reports automatically and allows quick generation of certain types of networks. With its simplified steps of execution, Saihu reduces the burden on users and makes it accessible for anyone working with time-sensitive networks.

cs.NI↗

arXiv2022

Equivalent Versions of Total Flow Analysis

Total Flow Analysis (TFA) is a method for conducting the worst-case analysis of time sensitive networks without cyclic dependencies. In networks with cyclic dependencies, Fixed-Point TFA introduces artificial cuts, analyses the resulting cycle-free network with TFA, and iterates. If it converges, it does provide valid performance bounds. We show that the choice of the specific cuts used by Fixed-Point TFA does not affect its convergence nor the obtained performance bounds, and that it can be replaced by an alternative algorithm that does not use any cut at all, while still applying to cyclic dependencies.

cs.NI↗