SEPIA Team

Head: Patricia STOLF

Most of the research works conducted in the SEPIA group address the issue of resource management in datacenters. In the following, we distinguish strategies which specifically target energy optimization in the datacenter (consumption and thermal effects), from works which address the improvement of virtualized datacenter consolidation (which is more general, but may also address energy saving). In a third category, we present works that focussed on the improvement of operating system support (at the level of a single server) in such environments.

Datacenter (energy) optimization

On the side of the environment of the datacenter, one research area is related to the cooling infrastructure and its optimisation. The challenge is to model the datacenter from a thermal perspective, considering the workload and application profiles, the power model and the cooling model. This model can be used to study thermal imbalances and to build strategies relying on physical  machine placement or task scheduling.

A second line of works at the infrastructure level is related to power provisioning with renewable energies. Integrating renewable energies to limit the CO2 emissions of datacenters attracted lots of attention in the last decade. The investigated approach is to influence scheduling in the datacenter according to the availability of renewable energies.

One of the issue with servers in Datacenters is that their power consumption is not fully proportional to the load, i.e. some idle power is consumed even when low load is experienced. We investigate a novel approach for building data centers with heterogeneous machines carefully chosen for their performance and energy efficiency ratios (Big, Medium, Little). Then a scheduler can exploit this infrastructure by migrating applications and switching machines on or off, so that the energy consumed by the data center is fully proportional to the load.

Consolidation of virtual machines, management of Cloud infrastructures

In virtualized datacenters, many optimization techniques are targeting server consolidation, i.e. packing virtual machines (VM) on as less servers as possible in order to power-off (suspend) unused servers. Several improvements to server consolidation can then be introduced.

One vein of improvement to consolidation is to adapt the architecture of the hosted applications in order to enable further consolidation, e.g. by changing the number of VMs or the VM sizes.

Another approach is to study heuristics which help improving VM placement, according to multiple criteria.

In cloud environments, the main limitation to consolidation is the lack of memory. Memory gets saturated way before CPU. We propose to exploit remote memory in the datacenter so that this constraint can be relaxed and consolidation improved.

OS support improvement, for Cloud and HPC

A significant part of our research targeted the improvement of operating system and virtualization supports used in datacenters.

An important cause of inefficiency in today’s virtualized environments is that scheduling decisions are made at two places, the hypervisor and the guest operating systems. A way of improvement is to establish a collaboration between these two levels.

One important issue of virtualization systems is to enforce predictability of performance. The performance of a guest operating system should not be perturbed by the fact that it executes in a virtual machine (VM). However, unpredictability can be observed for many reasons : cache contention, hypervisor level contention, hardware heterogeneity. We address the above sources of unpredictability.

Another important issue in virtualization is to take into account evolutions of hardware, and especially the shift towards NUMA architectures. In NUMA architectures, the major challenge comes from the fact that the hypervisor regularly reconfigures the placement of VM resources (vCPU, pages) over the NUMA topology. However, the guest operating systems are not designed to consider NUMA topology changes at runtime. This issue can be addressed at the different levels (application level, guest operating system level, hypervisor level).

One critical issue in virtualized systems is to have accurate monitoring tools regarding the behavior of VMs. This is a condition to implement efficient resource management policies. The issue is to implement monitoring tools which are not intruvive regarding the guest operating system or regarding performance, and which provide an accurate monitoring.

skills

distributed and autonomous systems
energy management
parallel optimization
cloud
data centers
virtualized operating systems

personnel of the team

Permanent members
Non-permanent members
Related members

publications of the team

International journals articles
National journals articles
Special issues of journal
International conferences articles
Conferences articles without published proceedings
Books Books edition Books parts
Thesis and HDR
Reports

contrats of the team

AcronymeTitreResp. scDébut – fin
i-Nondations anr i-Nondations : Gestion multi-vues des crues rapides Patricia STOLF
2017 – 2020
ENERGUMEN anr Optimiser l’énergie des plate-formes de calcul à large échelle Georges DA COSTA
2018 – 2022
SCALEVISOR Scalevisor: un hyperviseur pour rack désagrégé Daniel HAGIMONT
2018 – 2021
DATAZERO
[Contract completed]
anr Datacenters robustes utilisant de l’énergie renouvelable Jean-Marc PIERSON
2015 – 2020
Ctrl-Green
[Contract completed]
autre Administration autonome de Green Data Centers Daniel HAGIMONT
2012 – 2014
SOP
[Contract completed]
Penser service global pour les Ordinateurs Personnels Patricia STOLF
2011 – 2015
SelfXL
[Contract completed]
autre Administration autonome pour systèmes complexes de grande taille Daniel HAGIMONT
2008 – 2011
SATRIMMAP
[Contract completed]
autre Safety and time critical middleware for future modular avionics platform Emmanuel LAVINAL

Daniel HAGIMONT
2007 – 2010
AcronymeTitreResp. scDébut – fin
CoolEmAll
[Contract completed]
autre Platform for optimising the design and operation of modular configurable IT infrastructures and facilities with resource-efficient cooling Jean-Marc PIERSON
2011 – 2014
PIREGRID
[Contract completed]
autre Renforcer l’intégration transfrontalière en valorisant les complémentarités sur le plan des activités économiques, de l’innovation et du capital humain Jean-Marc PIERSON
2009 – 2012
AcronymeTitreResp. scDébut – fin
Blablamen
[Contract completed]
BlaBlaMem : un service de co-voiturage mémoire dans un datacenter Daniel HAGIMONT
2017 – 2019
2007-106
[Contract completed]
Administration d’applications scientifiques dans les grilles de calcul Daniel HAGIMONT
2007 – 2010
Monday 25 November 2019, 14h00
Scheduling in Cloud Data Center Powered by Renewable Energy Only With Mixed Phases-Based Workload
Gustavo ROSTIROLLA – Team SEPIA, IRIT UT3 Paul Sabatier, IRIT, Auditorium J. Herbrand
#these
Thursday 3 October 2019, 14h00
Gestion de centre de données avec sources d’énergies renouvelables intermittentes et incertaines
Léo GRANGE – Team SEPIA, IRIT UT3 Paul Sabatier, IRIT, Auditorium J. Herbrand
#these
Friday 14 June 2019, 14h00
Energy-efficient Resource Provisioning for Cloud Databases
Chaopeng GUO – Team SEPIA, IRIT UT3 Paul Sabatier, IRIT, Auditorium J. Herbrand
#these
Friday 28 September 2018, 14h30
Améliorer l’efficacité énergétique des datacenters virtualisés
Vlad NITU – Team SEPIA – IRIT INP-ENSEEIHT, Salle des thèses
#these
Monday 20 November 2017, 14h00
Towards a better use of renewable energies: application to scientific buildings
Inès DE COURCHELLE – Team SEPIA – IRIT UT3 Paul Sabatier, IRIT, Auditorium J. Herbrand
#these
Thursday 12 October 2017, 14h00
Performance et qualité de service de l’ordonnanceur dans un environnement virtualisé
Boris DJOMGWE TEABE – Team SEPIA – IRIT INP-ENSEEIHT, salle C002
#these
Tuesday 4 July 2017, 14h00
Self-Adaptive Bandwidth Control for Balanced QoS and Energy Aware Optimization in Wireless Sensor Network
Zong Yi LIU – Team SEPIA – IRIT INSA, Salle des thèses, Toulouse
#these
Friday 12 February 2016, 10h00
Data availability and security in cloud storage
Théodore Jean Richard RELAZA – Team SEPIA – IRIT UT3 Paul Sabatier, IRIT, Salle des Thèses
#these
Thursday 26 November 2015, 14h00
Contribution to the modeling of technology transfer in Green IT with Multi-Agent System
Christina HERZOG – Team SEPIA – IRIT UT3 Paul Sabatier, IRIT, Salle des Thèses
#these
Friday 17 July 2015, 14h00
Modeling the power consumption of computing systems and applications through Machine Learning techniques
Leandro FONTOURA CUPERTINO – Team SEPIA – IRIT UT3 Paul Sabatier, IRIT, Salle des Thèses
#these
Thursday 12 September 2019 – Friday 13 September 2019
Journées Cloud 2019 du GDR RSD (Réseau et Systèmes Distribués)
UT3 Paul Sabatier, IRIT, Auditorium J. Herbrand
#congres Know more
Tuesday 3 July 2018 – Friday 6 July 2018
Compas’2018 : Conférence d’informatique en Parallélisme, Architecture et Système
INP-ENSEEIHT
#congres Know more
Monday 2 July 2018 – Tuesday 3 July 2018
Edition 2018 des Green Days@Toulouse v2.0
INP-ENSEEIHT
#congres Know more
Monday 18 July 2016 – Thursday 21 July 2016
Workshop on Smart and Sustainable City
Bâtiment U4
#congres Know more
Thursday 12 November 2015 – Friday 13 November 2015
Journées Scientifiques SEPIA
UT3 Paul Sabatier, IRIT, Auditorium J. Herbrand
#congres Know more
Monday 16 March 2015 – Tuesday 17 March 2015
De l’énergétique au thermique, le Green IT est-il réellement vert ?
UT3 Paul Sabatier, IRIT
#congres Know more
Tuesday 3 March 2020, 9h30 – 10h30
Research problems in thread scheduling for multicore general purpose operating systems
Gilles MULLER – INRIA Paris UT3 Paul Sabatier, IRIT, Auditorium J. Herbrand
#seminaire
Friday 28 September 2018, 10h30 – 11h30
Enzian: making systems software research relevant in the face of custom hardware
Timothy ROSCOE – ETH (Suisse) INP-ENSEEIHT, Salle des thèses
#seminaire
Wednesday 11 July 2018, 15h00 – 16h30
From Pipes to Processors: Building High Performance Network Middleware
Timothy WOODS – George Washington University (Etats-Unis) INP-ENSEEIHT, Salle B307
#seminaire
Monday 20 November 2017, 10h00 – 12h00
Présentation du projet SeDuCe
Jean-Marc MENAUD – Ecole des Mines de Nantes UT3 Paul Sabatier, IRIT, Salle des Thèses
#seminaire
Friday 7 July 2017, 10h30 – 12h00
Everyday Computing with QCG MiddEveryday Computing with QCG Middleware and Toolsleware and Tools
Tomasz PIONTEK – Poznan Supercomputing and Networking Center (Pologne) UT3 Paul Sabatier, IRIT, Salle 175
#seminaire
Wednesday 2 July 2014, 10h00
Energy Efficient Computing from a Cloud Database Perspective
Jie SONG – Team SEPIA – IRIT (France) UT3 Paul Sabatier, IRIT, Salle des Thèses
#seminaire
Thursday 7 December 2017, 10h00
Resource management and performance improvement in virtualized infrastructures
Alain TCHANA – Team SEPIA – IRIT INP-ENSEEIHT, Salle des thèses
#hdr
Friday 13 November 2015, 10h00
La gestion des ressources pour des infrastructures vertes par la reconfiguration
Patricia STOLF – Team SEPIA – IRIT UT3 Paul Sabatier, IRIT, Auditorium J. Herbrand
#hdr
Thursday 12 November 2015, 14h30
Évaluation et optimisation de performance énergétique des centres de calcul
Georges DA COSTA – Team SEPIA – IRIT UT3 Paul Sabatier, IRIT, Auditorium J. Herbrand
#hdr
Thursday 11 December 2014, 10h00
Optimisation du stockage et de la gestion des données en environnements parallèles et distribués
Jacques JORDA – Team SEPIA – IRIT UT3 Paul Sabatier, IRIT, Salle des Thèses
#hdr