Thomas Tournaire - Academia.edu (original) (raw)

Uploads

Papers by Thomas Tournaire

Research paper thumbnail of Comparaisons de méthodes de calcul de seuils pour minimiser la consommation énergétique d’un cloud

ROADEF 2019: 20ème congrès annuel de la société Française de Recherche Opérationnelle et d’Aide à la Décision, Feb 19, 2019

Research paper thumbnail of Factored Reinforcement Learning for Auto-scaling in Tandem Queues

NOMS 2022-2022 IEEE/IFIP Network Operations and Management Symposium

Research paper thumbnail of Génération de seuils optimaux dans une file hystérésis : application à un modèle de cloud

Nous proposons dans cette etude des methodes d'optimisation efficaces pour la recherche de va... more Nous proposons dans cette etude des methodes d'optimisation efficaces pour la recherche de valeurs de seuils dans une file d'attente multi-serveur hysteresis. Il s'agit de minimiser un cout global prenant en compte les performances et l'utilisation des ressources et se calculant comme une fonction de cout sur la distribution stationnaire. Ce cout global etant une fonction non convexe, la recherche de la valeur optimale est complexe. Nous proposons deux types de methodes d'optimisation : l'une est basee sur des heuristiques, couplees a de l'agregation de chaines de Markov pour reduire les temps d'execution et l'autre est une meta-heuristique plus precisement le recuit simule

Research paper thumbnail of Optimal control policies for resource allocation in the Cloud: comparison between Markov decision process and heuristic approaches

ArXiv, 2021

We consider an auto-scaling technique in a cloud system where virtual machines hosted on a physic... more We consider an auto-scaling technique in a cloud system where virtual machines hosted on a physical node are turned on and off depending on the queue’s occupation (or thresholds), in order to minimise a global cost integrating both energy consumption and performance. We propose several efficient optimisation methods to find threshold values minimising this global cost: local search heuristics coupled with aggregation of Markov chain and with queues approximation techniques to reduce the execution time and improve the accuracy. The second approach tackles the problem with a Markov Decision Process (MDP) for which we proceed to a theoretical study and provide theoretical comparison with the first approach. We also develop structured MDP algorithms integrating hysteresis properties. We show that MDP algorithms (value iteration, policy iteration) and especially structured MDP algorithms outperform the devised heuristics, in terms of time execution and accuracy. Finally, we propose a cos...

Research paper thumbnail of Generating Optimal Thresholds in a Hysteresis Queue: Application to a Cloud Model

Reducing the energy consumption of a cloud system while guaranteeing a given quality of service l... more Reducing the energy consumption of a cloud system while guaranteeing a given quality of service level is a crucial problem encountered today by cloud providers. We consider an auto-scaling model where virtual machines are turned on and off depending on the queue's occupation (or thresholds). This model represents the variability of allocated resources (Virtual Machines or VMs) according to user demands. It can be studied using an hysteresis queuing model, which is represented by a multidimensional Markov chain, whose calculation of the stationary distribution becomes complex when the number of VMs grows. We adopt a cost-aware approach and define a mean cost computed as a reward function on the stationary distribution. This cost takes into account both the performance (for Service Level Agreement: SLA) and the use of the resources (for Energy). We propose efficient optimisation methods to find threshold values minimising the global cost. Because this mean cost is a non-convex fun...

Research paper thumbnail of Reinforcement Learning with Model-Based Approaches for Dynamic Resource Allocation in a Tandem Queue

Lecture Notes in Computer Science

Research paper thumbnail of Comparaisons de méthodes de calcul de seuils pour minimiser la consommation énergétique d’un cloud

ROADEF 2019: 20ème congrès annuel de la société Française de Recherche Opérationnelle et d’Aide à la Décision, Feb 19, 2019

Research paper thumbnail of Factored Reinforcement Learning for Auto-scaling in Tandem Queues

NOMS 2022-2022 IEEE/IFIP Network Operations and Management Symposium

Research paper thumbnail of Génération de seuils optimaux dans une file hystérésis : application à un modèle de cloud

Nous proposons dans cette etude des methodes d'optimisation efficaces pour la recherche de va... more Nous proposons dans cette etude des methodes d'optimisation efficaces pour la recherche de valeurs de seuils dans une file d'attente multi-serveur hysteresis. Il s'agit de minimiser un cout global prenant en compte les performances et l'utilisation des ressources et se calculant comme une fonction de cout sur la distribution stationnaire. Ce cout global etant une fonction non convexe, la recherche de la valeur optimale est complexe. Nous proposons deux types de methodes d'optimisation : l'une est basee sur des heuristiques, couplees a de l'agregation de chaines de Markov pour reduire les temps d'execution et l'autre est une meta-heuristique plus precisement le recuit simule

Research paper thumbnail of Optimal control policies for resource allocation in the Cloud: comparison between Markov decision process and heuristic approaches

ArXiv, 2021

We consider an auto-scaling technique in a cloud system where virtual machines hosted on a physic... more We consider an auto-scaling technique in a cloud system where virtual machines hosted on a physical node are turned on and off depending on the queue’s occupation (or thresholds), in order to minimise a global cost integrating both energy consumption and performance. We propose several efficient optimisation methods to find threshold values minimising this global cost: local search heuristics coupled with aggregation of Markov chain and with queues approximation techniques to reduce the execution time and improve the accuracy. The second approach tackles the problem with a Markov Decision Process (MDP) for which we proceed to a theoretical study and provide theoretical comparison with the first approach. We also develop structured MDP algorithms integrating hysteresis properties. We show that MDP algorithms (value iteration, policy iteration) and especially structured MDP algorithms outperform the devised heuristics, in terms of time execution and accuracy. Finally, we propose a cos...

Research paper thumbnail of Generating Optimal Thresholds in a Hysteresis Queue: Application to a Cloud Model

Reducing the energy consumption of a cloud system while guaranteeing a given quality of service l... more Reducing the energy consumption of a cloud system while guaranteeing a given quality of service level is a crucial problem encountered today by cloud providers. We consider an auto-scaling model where virtual machines are turned on and off depending on the queue's occupation (or thresholds). This model represents the variability of allocated resources (Virtual Machines or VMs) according to user demands. It can be studied using an hysteresis queuing model, which is represented by a multidimensional Markov chain, whose calculation of the stationary distribution becomes complex when the number of VMs grows. We adopt a cost-aware approach and define a mean cost computed as a reward function on the stationary distribution. This cost takes into account both the performance (for Service Level Agreement: SLA) and the use of the resources (for Energy). We propose efficient optimisation methods to find threshold values minimising the global cost. Because this mean cost is a non-convex fun...

Research paper thumbnail of Reinforcement Learning with Model-Based Approaches for Dynamic Resource Allocation in a Tandem Queue

Lecture Notes in Computer Science

Log In