Vincent Kherbache

I am a PhD student in computer science at INRIA Sophia Antipolis where I'm part of the SCALE team since september 2013.

Learn more about what I do

Work and research interests

The topic of my thesis consists of improving the internal management of data centres by reducing the energy footprint of common maintenance operations while optimizing the use of renewable-energy. The underlying objective is also to simplify the management of data centres, and therefore the work of operators, by maintaining a simple declarative approach to perform complex maintenance tasks.

More generally, my research interests focus on virtualized resources management, adaptive maintenance operations and energy-efficiency in renewable-powered Data Centres.

Cloud computing

Renewable energy

Data Centre mgmt

My thesis is mainly founded by the European 7th FP project DC4Cities, whose main objective is to let existing and new data centres become energy adaptive.



  • 2013 - Master's Degree in Network Computing and Embedded Systems (RISE) at University of Strasbourg (France).
  • 2011 - Bachelor's Degree in Computer Science at University Paris-Est Créteil (France).
  • 2010 - Graduated with a Professionnal Bachelor degree in Computer Science (ASRALL) at University Nancy 2 (France).
  • 2009 - Graduated with an advanced vocational diploma in Computer Science.
  • 2007 - High School diploma in Electrical Engineering.

Full list of publications

Scheduling Live-Migrations for Fast, Adaptable and Energy Efficient Relocation Operations. [Paper] [Slides]

Authors: Vincent Kherbache, Fabien Hermenier, Éric Madelaine

Conference paper: UCC'15 - 8th IEEE/ACM International Conference on Utility and Cloud Computing. December 2015. Limassol, Cyprus.

Abstract: Every day, numerous VMs are migrated inside a datacenter to balance the load, save energy or prepare production servers for maintenance. Despite VM placement problems are carefully studied, the underlying migration scheduler rely on vague adhoc models. This leads to unnecessarily long and energy-intensive migrations.
We present mVM, a new and extensible migration scheduler. mVM takes into account the VM memory workload and the network topology to estimate precisely the migration duration and take wiser scheduling decisions. mVM is implemented as a plugin of BtrPlace and can be customized with additional scheduling constraints to finely control the migrations. Experiments on a real testbed show mVM outperforms schedulers that cap the migration parallelism by a constant to reduce the completion time. Besides an optimal capping, mVM reduces the migration duration by 20.4% on average and the completion time by 28.1%. In a maintenance operation involving 96 VMs to migrate between 72 servers, mVM saves 21.5% Joules against BtrPlace. Finally, its current library of 6 constraints allows administrators to address temporal and energy concerns, for example to adapt the schedule and fit a power budget.

Ordonnancement contrôlé de migrations à chaud. [Paper] [Slides]

Authors: Vincent Kherbache, Fabien Hermenier, Éric Madelaine

French paper: Compas'15 - Conférence d’informatique en Parallélisme, Architecture et Système. July 2015. Lille, France.

Abstract: Migrer à chaud une machine virtuelle (VM) est une opération basique dans un centre de données. Tous les jours, des VM sont migrées pour répartir la charge, économiser de l’énergie ou préparer la maintenance de serveurs en production. Bien que les problèmes de placement des VM soient beaucoup étudiés, on observe que la gestion des migrations permettant de transiter vers ces nouveaux placements reste un domaine de second plan. On observe alors des algorithmes de placement de qualité, couplés à des algorithmes d’ordonnancement prenant des décisions peu pertinentes causées par des hypothèses irréalistes.
Nous présentons dans ce papier mVM, un ordonnanceur de migrations reposant sur un modèle précis du réseau et du protocole de migration à chaud. Cet ordonnanceur a été intégré en place de celui du gestionnaire de VM BtrPlace. Nos premières expérimentations montrent que les durées des migrations sont estimées à 1.5 secondes près. Cette précision a permis de calculer de meilleurs ordonnancements, réduisant la durée des migrations par 3.5 comparée à BtrPlace.

Memory and Network Aware Scheduling of Virtual Machine Migrations. [PDF]

Authors: Vincent Kherbache, Fabien Hermenier, Éric Madelaine

Poster: EuroSys'15 - European Conference on Computer Systems - Poster session. April 2015. Bordeaux, France. ACM.

Dynamic Packing with Side Constraints for Datacenter Resource Management. [Paper]

Authors: Sophie Demassey, Fabien Hermenier, Vincent Kherbache

Book chapter: Optimized Packings and Their Applications, Springer, 2015.

Abstract: Resource management in datacenters involves assigning virtual machines with changing resource demands to physical machines with changing capacities. Recurrently, the changes invalidate the assignment and the resource manager recomputes it at runtime. The assignment is also subject to changing restrictions expressing a variety of user requirements. The present chapter surveys this application of vector packing -- called the VM reassignment problem -- with an insight into its dynamic and heterogeneous nature. We advocate flexibility to answer these issues and present BtrPlace, a flexible and scalable heuristic solution based on Constraint Programming.

Planning live-migrations to prepare servers for maintenance. [Paper] [Slides]

Authors: Vincent Kherbache, Fabien Hermenier, Éric Madelaine

Workshop paper: VHPC'14 - 9th Workshop on Virtualization in High-Performance Cloud Computing. August 2014. Porto, Portugal. ACM.

Abstract: In a virtualized data center, server maintenance is a common but still critical operation. A prerequisite is indeed to relocate elsewhere the Virtual Machines (VMs) running on the production servers to prepare them for the maintenance. When the maintenance focuses several servers, this may lead to a costly relocation of several VMs so the migration plan must be chose wisely. This however implies to master numerous human, technical, and economical aspects that play a role in the design of a quality migration plan. In this paper, we study migration plans that can be decided by an operator to prepare for an hardware upgrade or a server refresh on multiple servers. We exhibit performance bottleneck and pitfalls that reduce the plan efficiency. We then discuss and validate possible improvements deduced from the knowledge of the environment peculiarities.

Contact informations

vincent.kherbache |4T| inria |D0T| fr

GPG key: 91009937

vincent.kherbache |4T| gmail |D0T| com