Dimitri PERTIN

Département

Département Automatique Productique Informatique

Localisation

Nantes

Coordonnées :

Publications HAL
Communication dans un congrès
Cherrueau Ronan-Alexandre, Lèbre Adrien, Pertin Dimitri, Wuhib Fetahi, Soares João
Edge Computing Resource Management System: a Critical Building Block! Initiating the debate via OpenStack
HotEdge 2018 - USENIX Workshop on Hot Topics in Edge Computing, Jul 2018, Boston, MA, United States. pp.1-6
Bibtext :
@inproceedings{cherrueau:hal-01812747,
TITLE = {{Edge Computing Resource Management System: a Critical Building Block! Initiating the debate via OpenStack}},
AUTHOR = {Cherrueau, Ronan-Alexandre and L{\`e}bre, Adrien and Pertin, Dimitri and Wuhib, Fetahi and Soares, João Monteiro},
URL = {https://hal.archives-ouvertes.fr/hal-01812747},
BOOKTITLE = {{HotEdge 2018 - USENIX Workshop on Hot Topics in Edge Computing}},
ADDRESS = {Boston, MA, United States},
PAGES = {1-6},
YEAR = {2018},
MONTH = Jul,
PDF = {https://hal.archives-ouvertes.fr/hal-01812747/file/hotedge2018.pdf},
HAL_ID = {hal-01812747},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Proceedings
%T Edge Computing Resource Management System: a Critical Building Block! Initiating the debate via OpenStack
%+ Software Stack for Massively Geo-Distributed Infrastructures (STACK)
%+ Automatique, productique et informatique (DAPI)
%+ Laboratoire des Sciences du Numérique de Nantes (LS2N)
%+ Aspect and Composition Languages (ASCOLA)
%+ Ericsson Research
%A Cherrueau, Ronan-Alexandre
%A Lèbre, Adrien
%A Pertin, Dimitri
%A Wuhib, Fetahi
%A Soares, João, Monteiro
%< avec comité de lecture
%B HotEdge 2018 - USENIX Workshop on Hot Topics in Edge Computing
%C Boston, MA, United States
%P 1-6
%8 2018-07-10
%D 2018
%Z Computer Science [cs]/Operating Systems [cs.OS]
%Z Computer Science [cs]/Software Engineering [cs.SE]
%Z Computer Science [cs]/General Literature [cs.GL]Conference papers
%X While it is clear that edge infrastructures are required for emerging use-cases related to IoT, VR or NFV, there is currently no resource management system able to deliver all features for the edge that made cloud computing successful (e.g., an OpenStack for the edge). Since building a system from scratch is seen by many as impractical, this paper provides reflections regarding how existing solutions can be leveraged. To that end, we provide a list of the features required to operate and use edge computing resources, and investigate how an existing IaaS manager (i.e., OpenStack) satisfies these requirements. Finally, we identify from this study two approaches to design an edge infrastructure manager that fulfils our requirements, and discuss their pros and cons.
%G English
%2 https://hal.archives-ouvertes.fr/hal-01812747/document
%2 https://hal.archives-ouvertes.fr/hal-01812747/file/hotedge2018.pdf
%L hal-01812747
%U https://hal.archives-ouvertes.fr/hal-01812747
%~ CNRS
%~ EC-NANTES
%~ INRIA-RENNES
%~ UNAM
%~ INRIA_TEST
%~ LS2N
%~ INRIA
%~ LS2N-STACK
%~ LS2N-STACK-IMTA
%~ IMTA_DAPI
%~ UNIV-NANTES
%~ INRIA-RENGRE
%~ IMT-ATLANTIQUE
Communication dans un congrès
Coullon Hélène, Pertin Dimitri, Pérez Christian
Production Deployment Tools for IaaSes: an Overall Model and Survey
The IEEE 5th International Conference on Future Internet of Things and Cloud (FiCloud), Aug 2017, Prague, Czech Republic. pp.183-190, 2017, 〈http://www.ficloud.org/2017/〉. 〈10.1109/FiCloud.2017.51〉
Bibtext :
@inproceedings{coullon:hal-01532489,
TITLE = {{Production Deployment Tools for IaaSes: an Overall Model and Survey}},
AUTHOR = {Coullon, H{\'e}l{\`e}ne and Pertin, Dimitri and P{\'e}rez, Christian},
URL = {https://hal.inria.fr/hal-01532489},
BOOKTITLE = {{The IEEE 5th International Conference on Future Internet of Things and Cloud (FiCloud)}},
ADDRESS = {Prague, Czech Republic},
PAGES = {183-190},
YEAR = {2017},
MONTH = Aug,
DOI = {10.1109/FiCloud.2017.51},
KEYWORDS = { Survey ; Deployment ; Openstack ; IaaS ; Model},
PDF = {https://hal.inria.fr/hal-01532489/file/main.pdf},
HAL_ID = {hal-01532489},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Proceedings
%T Production Deployment Tools for IaaSes: an Overall Model and Survey
%+ Automatique, productique et informatique (DAPI)
%+ Aspect and Composition Languages (ASCOLA)
%+ Laboratoire des Sciences du Numérique de Nantes (LS2N)
%+ Algorithms and Software Architectures for Distributed and HPC Platforms (AVALON)
%A Coullon, Hélène
%A Pertin, Dimitri
%A Pérez, Christian
%Z IPL Discovery
%< avec comité de lecture
%B The IEEE 5th International Conference on Future Internet of Things and Cloud (FiCloud)
%C Prague, Czech Republic
%P 183-190
%8 2017-08-21
%D 2017
%R 10.1109/FiCloud.2017.51
%K Survey
%K Deployment
%K Openstack
%K IaaS
%K Model
%Z Computer Science [cs]
%Z Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC]
%Z Computer Science [cs]/Modeling and SimulationConference papers
%X Emerging applications for the Internet of Things (IoT) are complex programs which are composed of multiple modules (or services). For scalability, reliability and performance, modular applications are distributed on infrastructures that support utility computing (e.g., Cloud, Fog). In order to simply operate such infrastructures, an Infrastructure-as-a-Service (IaaS) manager is required. OpenStack is the de-facto open-source solution to address the IaaS level of the Cloud paradigm. However, OpenStack is itself a large modular application composed of more than 150 modules that make it hard to deploy manually. To fully understand how IaaSes are deployed today, we propose in this paper an overall model of the application deployment process which describes each step with their interactions. This model then serves as the basis to analyse five different deployment tools used to deploy OpenStack in production: Kolla, Enos, Juju, Kubernetes, and TripleO. Finally, a comparison is provided and the results are discussed to extend this analysis.
%G English
%2 https://hal.inria.fr/hal-01532489/document
%2 https://hal.inria.fr/hal-01532489/file/main.pdf
%L hal-01532489
%U https://hal.inria.fr/hal-01532489
%~ CNRS
%~ EC-NANTES
%~ INRIA-RHA
%~ UNAM
%~ TDS-MACS
%~ LIP
%~ UNIV-NANTES
%~ ENS-LYON
%~ INRIA_TEST
%~ INRIA2
%~ LS2N
%~ INRIA2017
%~ INRIA
%~ LS2N-ASCOLA
%~ UNIV-LYON1
%~ LS2N-STACK
%~ IMTA_DAPI
%~ INRIA-RENGRE
%~ INRIA-RENNES
%~ IMT-ATLANTIQUE
Communication dans un congrès
Cherrueau Ronan-Alexandre, Pertin Dimitri, Simonet Anthony, Lebre Adrien, Simonin Matthieu
Toward a Holistic Framework for Conducting Scientific Evaluations of OpenStack
The 17th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID), May 2017, Madrid, Spain. The 17th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID), pp.544-548, 2017, 〈https://www.arcos.inf.uc3m.es/ccgrid2017/〉. 〈10.1109/CCGRID.2017.87〉
Bibtext :
@inproceedings{cherrueau:hal-01664515,
TITLE = {{Toward a Holistic Framework for Conducting Scientific Evaluations of OpenStack}},
AUTHOR = {Cherrueau, Ronan-Alexandre and Pertin, Dimitri and Simonet, Anthony and Lebre, Adrien and Simonin, Matthieu},
URL = {https://hal.archives-ouvertes.fr/hal-01664515},
BOOKTITLE = {{The 17th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID)}},
ADDRESS = {Madrid, Spain},
PAGES = {544-548},
YEAR = {2017},
MONTH = May,
DOI = {10.1109/CCGRID.2017.87},
KEYWORDS = { Measurement ; Cloud computing ; Benchmark testing ; Containers ; Tools ; Topology },
PDF = {https://hal.archives-ouvertes.fr/hal-01664515/file/main.pdf},
HAL_ID = {hal-01664515},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Proceedings
%T Toward a Holistic Framework for Conducting Scientific Evaluations of OpenStack
%+ Aspect and Composition Languages (ASCOLA)
%+ Automatique, productique et informatique (DAPI)
%+ Laboratoire des Sciences du Numérique de Nantes (LS2N)
%+ Design and Implementation of Autonomous Distributed Systems (MYRIADS)
%A Cherrueau, Ronan-Alexandre
%A Pertin, Dimitri
%A Simonet, Anthony
%A Lebre, Adrien
%A Simonin, Matthieu
%Z IPL Discovery
%< avec comité de lecture
%( The 17th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID)
%B The 17th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID)
%C Madrid, Spain
%P 544-548
%8 2017-05-14
%D 2017
%R 10.1109/CCGRID.2017.87
%K Measurement
%K Cloud computing
%K Benchmark testing
%K Containers
%K Tools
%K Topology
%Z Computer Science [cs]
%Z Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC]
%Z Computer Science [cs]/Performance [cs.PF]Conference papers
%X By massively adopting OpenStack for operating small to large private and public clouds, the industry has made it one of the largest running software project, overgrowing the Linux kernel. However, with success comes increased complexity; facing technical and scientific challenges , developers are in great difficulty when testing the impact of individual changes on the performance of such a large codebase, which will likely slow down the evolution of OpenStack. Thus, we claim it is now time for the scientific community to join the effort and get involved in the development of OpenStack, like it has been once done for Linux. In this spirit, we developed Enos, an integrated framework that relies on container technologies for deploying and evaluating OpenStack on any testbed. Enos allows researchers to easily express different configurations, enabling fine-grained investigations of OpenStack services. Enos collects performance metrics at runtime and stores them for post-mortem analysis and sharing. The relevance of the Enos approach to reproducible research is illustrated by evaluating different OpenStack scenarios on the Grid'5000 testbed.
%G English
%2 https://hal.archives-ouvertes.fr/hal-01664515/document
%2 https://hal.archives-ouvertes.fr/hal-01664515/file/main.pdf
%L hal-01664515
%U https://hal.archives-ouvertes.fr/hal-01664515
%~ CNRS
%~ UNIV-UBS
%~ EC-NANTES
%~ IRISA_SET
%~ UNAM
%~ UNIV-NANTES
%~ UNIV-RENNES1
%~ INRIA_TEST
%~ LORIA2
%~ INRIA
%~ UR1-HAL
%~ IRISA
%~ INRIA2017
%~ LS2N
%~ UR1-MATH-STIC
%~ UR1-UFR-ISTIC
%~ LS2N-ASCOLA
%~ GRID5000
%~ LS2N-STACK
%~ TEST-UNIV-RENNES
%~ TEST-UR-CSS
%~ IMTA_DAPI
%~ UNIV-RENNES
%~ INRIA-RENNES
%~ INRIA-RENGRE
%~ CENTRALESUPELEC
%~ IMT-ATLANTIQUE
Rapport
Cherrueau Ronan-Alexandre, Lebre Adrien, Pertin Dimitri, Simonet Anthony, Simonin Matthieu
ENOS: a Holistic Framework for Conducting Scientific Evaluations of OpenStack
[Technical Report] RT-0485, Inria Rennes Bretagne Atlantique; Nantes. 2016
Bibtext :
@techreport{cherrueau:hal-01415522,
TITLE = {{ENOS: a HolisticFramework forConducting ScientificEvaluations of OpenStack}},
AUTHOR = {Cherrueau, Ronan-Alexandre and Lebre, Adrien and Pertin, Dimitri and Simonet, Anthony and Simonin, Matthieu},
URL = {https://hal.inria.fr/hal-01415522},
TYPE = {Technical Report},
NUMBER = {RT-0485},
INSTITUTION = {{Inria Rennes Bretagne Atlantique ; Nantes}},
YEAR = {2016},
MONTH = Nov,
KEYWORDS = { Cloud ; Visualization ; Performance ; Data-plane ; Control-plane ; Reproducibility ; OpenStack},
PDF = {https://hal.inria.fr/hal-01415522/file/RT-485.pdf},
HAL_ID = {hal-01415522},
HAL_VERSION = {v2},
}
Endnote :
%0 Report
%T ENOS: a HolisticFramework forConducting ScientificEvaluations of OpenStack
%+ Aspect and composition languages (ASCOLA)
%+ Design and Implementation of Autonomous Distributed Systems (MYRIADS)
%A Cherrueau, Ronan-Alexandre
%A Lebre, Adrien
%A Pertin, Dimitri
%A Simonet, Anthony
%A Simonin, Matthieu
%Z IPL Discovery
%N RT-0485
%I Inria Rennes Bretagne Atlantique
%I Nantes
%8 2016-11-30
%D 2016
%K Cloud
%K Visualization
%K Performance
%K Data-plane
%K Control-plane
%K Reproducibility
%K OpenStack
%Z Computer Science [cs]/Computation and Language [cs.CL]Reports
%X By massively adopting OpenStack for operating small to large private and public clouds, the industry has made it one of the largest running software project. Driven by an incredibly vibrant community, OpenStack has now overgrown the Linux kernel. However, with success comes an increased complexity; facing technical and scientific challenges, developers are in great difficulty when testing the impact of individual changes on the performance of such a large codebase, which will likely slow down the evolution of OpenStack. In the light of the difficulties the OpenStack community is facing, we claim that it is time for our scientific community to join the effort and get involved in the development and the evolution of OpenStack, as it has been once done for Linux. However, diving into complex software such as OpenStack is tedious: reliable tools are necessary to ease the efforts of our community and make science as collaborative as possible.In this spirit, we developed ENOS, an integrated framework that relies on container technologies for deploying and evaluating OpenStack on any testbed. ENOS allows researchers to easily express different configurations, enabling fine-grained investigations of OpenStack services. ENOS collects performance metrics at runtime and stores them for post-mortem analysis and sharing. The relevance of ENOS approach to reproducible research is illustrated by evaluating different OpenStack scenarios on the Grid’5000 testbed.
%G English
%Z Discovery
%2 https://hal.inria.fr/hal-01415522v2/document
%2 https://hal.inria.fr/hal-01415522/file/RT-485.pdf
%L hal-01415522
%U https://hal.inria.fr/hal-01415522
%~ CNRS
%~ UNIV-UBS
%~ LINA
%~ INSTITUT-TELECOM
%~ IRISA_SET
%~ UNIV-NANTES
%~ UNIV-RENNES1
%~ INRIA
%~ IRISA
%~ INRIA_TEST
%~ INRIA2
%~ LORIA2
%~ DAPI
%~ LARA
%~ MINES-NANTES
%~ UR1-HAL
%~ UR1-MATH-STIC
%~ INRIA-RRRT
%~ UR1-UFR-ISTIC
%~ GRID5000
%~ LINA-ASCOLA
%~ TEST-UNIV-RENNES
%~ TEST-UR-CSS
%~ IMTA_DAPI
%~ UNIV-RENNES
%~ INRIA-RENNES
%~ INRIA-RENGRE
%~ IMT-ATLANTIQUE
Thèse
Pertin Dimitri
Code à Effacement Mojette pour le Stockage Distribué
Réseaux et télécommunications [cs.NI]. Université de Nantes, 2016. Français
Bibtext :
@phdthesis{pertin:tel-01720038,
TITLE = {{Mojette Erasure Code for Distributed Storage}},
AUTHOR = {Pertin, Dimitri},
URL = {https://hal.archives-ouvertes.fr/tel-01720038},
SCHOOL = {{Universit{\'e} de Nantes}},
YEAR = {2016},
MONTH = Apr,
KEYWORDS = { fault tolerance ; distributed storage ; Mojette transform ; Erasure code ; tol{\'e}rance aux pannes ; stockage distribu{\'e} ; transform{\'e}e Mojette ; Code {\`a} effacement},
TYPE = {Theses},
PDF = {https://hal.archives-ouvertes.fr/tel-01720038/file/pertin_these.pdf},
HAL_ID = {tel-01720038},
HAL_VERSION = {v1},
}
Endnote :
%0 Thesis
%T Mojette Erasure Code for Distributed Storage
%+ Institut de Recherche en Communications et en Cybernétique de Nantes (IRCCyN)
%+ Rozo Systems
%A Pertin, Dimitri
%Z CIFRE
%I Université de Nantes
%Y Nicolas Normand
%Z M. Imants Svalbe, Senior Lecturer, Université de Monash, Melbourne, Australie - Président du jury
%Z M. Pierre Duhamel, Directeur de recherche au CNRS, L2S, Centrale Supélec, Paris - Rapporteur
%Z M. Martin Quinson, Professeur des universités, ENS, Rennes - Rapporteur
%Z M. Jérôme Lacan, Professeur, ISAE-SUPAERO, Toulouse - Examinateur
%Z M. Pierre Évenou, Ingénieur, Rozo Systems, Nantes - Invité
%Z M. Nicolas Normand, Maître de conférences titulaire de l'HDR, Université de Nantes - Directeur de thèse
%Z M. Benoît Parrein, Maître de conférences titulaire de l'HDR, Université de Nantes - Co-encadrant de thèse
%8 2016-04-22
%D 2016
%K fault tolerance
%K distributed storage
%K Mojette transform
%K Erasure code
%K tolérance aux pannes
%K stockage distribué
%K transformée Mojette
%K Code à effacement
%Z Computer Science [cs]/Networking and Internet Architecture [cs.NI]
%Z Computer Science [cs]/Discrete Mathematics [cs.DM]
%Z Computer Science [cs]/Computational Geometry [cs.CG]
%Z Computer Science [cs]/Information Theory [cs.IT]
%Z Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC]
%Z Computer Science [cs]/Multiagent Systems [cs.MA]Theses
%X Erasure codes can generate data redundancy in distributed storage systems. This redundancy can be used to recover missing data in case of a failure. Codes have the benefit of reducing the generated amount of redundancy drastically,compared to plain data replication. However, this reduction is combined with a significant computational complexity, which penalizes encoding and decoding performances, and limits the use of coding to cold data.In this thesis, we focus on the use of the Mojette transform as an effective erasure code, adapted to hot data. The resulting code requires more redundancy than classicalcodes though.The first contribution of this research work deals with the design of a systematic version of the Mojette erasure code. This version provides better performances while reducing the required amount of redundancy.The second contribution covers the integration of this solution in the distributed file system RozoFS. This integration enables the system to provide a continuous service despite failures, while being able to manage hot data with half the volume of data compared to replication-basedsystems.A third research focus addresses the design of a distributed method to compute extra Mojette codeword symbols. This method contributes to restore a redundancy threshold in thestorage system.
%X Les codes à effacement permettent de générer de la redondance de données numériques dans un système de stockage distribué. Cette redondance permet de restaurer une partie manquante des données en cas de panne.L’avantage des codes est de réduire considérablement la quantité de redondance générée par rapport aux techniques classiques de réplication. Toutefois, cette réduction s’accompagne d’une complexité calculatoire significative, pénalisant les performances d’encodage et de décodage, cequi limite leur utilisation aux données froides. Dans cette thèse, nous nous intéressons à l’utilisation de la transformation Mojette afin de fournir un code à effacement performant, adapté aux données chaudes. Le code qui en résulte nécessite cependant plus de redondance par rapportaux codes classiques.La première contribution de ces travaux de thèse traite de la conception d’une version systématique du code à effacement Mojette. Cette version a l’avantage d’augmenter significativement les performances du code, tout en réduisant la quantité de redondance nécessaire.La seconde contribution s’intéresse à l’intégration de cette solution au sein du système de fichiers distribué RozoFS. Cette contribution permet au système d’assurer un service continu en cas de panne, tout en étant capable de gérer les données chaudes avec deux fois moins de données par rapport aux systèmes basés sur la réplication.Un troisième axe de recherche se focalise sur la conception d’une méthode distribuée pour générer de nouveaux symboles de mots de code Mojette. Cette technique participe à la restauration d’un seuil de redondance du système de stockage.
%G French
%2 https://hal.archives-ouvertes.fr/tel-01720038/document
%2 https://hal.archives-ouvertes.fr/tel-01720038/file/pertin_these.pdf
%L tel-01720038
%U https://hal.archives-ouvertes.fr/tel-01720038
%~ IRCCYN
%~ CNRS
%~ EC-NANTES
%~ UNAM
%~ TDS-MACS
%~ UNIV-NANTES-THESES
%~ UNIV-NANTES
%~ GRID5000
Communication dans un congrès
Pertin Dimitri, Van Kempen Alexandre, Parrein Benoît, Normand Nicolas
Comparison of RAID-6 Erasure Codes
The third Sino-French Workshop on Information and Communication Technologies, SIFWICT 2015, Jun 2015, Nantes, France
Bibtext :
@inproceedings{pertin:hal-01162047,
TITLE = {{Comparison of RAID-6 Erasure Codes}},
AUTHOR = {Pertin, Dimitri and Van Kempen, Alexandre and Parrein, Beno{\^i}t and Normand, Nicolas},
URL = {https://hal.archives-ouvertes.fr/hal-01162047},
BOOKTITLE = {{The third Sino-French Workshop on Information and Communication Technologies, SIFWICT 2015}},
ADDRESS = {Nantes, France},
YEAR = {2015},
MONTH = Jun,
KEYWORDS = {Erasure Coding ; Storage ; Mojette Transform},
PDF = {https://hal.archives-ouvertes.fr/hal-01162047/file/pertin2015sifwict.pdf},
HAL_ID = {hal-01162047},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Paper
%F Oral
%T Comparison of RAID-6 Erasure Codes
%+ Institut de Recherche en Communications et en Cybernétique de Nantes (IRCCyN)
%+ Rozo Systems
%A Pertin, Dimitri
%A Van Kempen, Alexandre
%A Parrein, Benoît
%A Normand, Nicolas
%< avec comité de lecture
%B The third Sino-French Workshop on Information and Communication Technologies, SIFWICT 2015
%C Nantes, France
%8 2015-06-12
%D 2015
%K Erasure Coding
%K Storage
%K Mojette Transform
%Z Computer Science [cs]
%Z Computer Science [cs]/Networking and Internet Architecture [cs.NI]
%Z Computer Science [cs]/Performance [cs.PF]Conference papers
%X Coding techniques for RAID-6 storage systems, providing a doublefault-tolerance, are varied. They all come with their respective benefits andlimitations. In this paper, we compare the characteristics of five prominenterasure codes. We show that the general-purpose Reed-Solomon codes, eitherbased on Vandermonde or Cauchy matrices, are outperformed by Array codes (i.e.EVENODD, RDP) which are specifically designed for RAID-6 storage systems.However geometrical codes such as the Mojette erasure code show evenbetter performances at the cost of a slight extra storage overhead. We outlinethe differences between these codes in terms of encoding, updating and decodingcomplexity. We believe that such an analysis can be valuable to systemdesigners to figure out which code would best suit their requirements.
%G English
%2 https://hal.archives-ouvertes.fr/hal-01162047/document
%2 https://hal.archives-ouvertes.fr/hal-01162047/file/pertin2015sifwict.pdf
%L hal-01162047
%U https://hal.archives-ouvertes.fr/hal-01162047
%~ UNIV-NANTES
%~ CNRS
%~ EC-NANTES
%~ IRCCYN
%~ IRCCYN-IVC
%~ UNAM
%~ IVC-IPI
Pré-publication, Document de travail
Pertin Dimitri, Féron Didier, Van Kempen Alexandre, Parrein Benoît
Performance evaluation of the Mojette erasure code for fault-tolerant distributed hot data storage
2015
Bibtext :
@unpublished{pertin:hal-01156443,
TITLE = {{Performance evaluation of the Mojette erasure code for fault-tolerant distributed hot data storage}},
AUTHOR = {Pertin, Dimitri and F{\'e}ron, Didier and Van Kempen, Alexandre and Parrein, Beno{\^i}t},
URL = {https://hal.archives-ouvertes.fr/hal-01156443},
NOTE = {working paper or preprint},
YEAR = {2015},
MONTH = May,
PDF = {https://hal.archives-ouvertes.fr/hal-01156443/file/hotstorage15-paper25.pdf},
HAL_ID = {hal-01156443},
HAL_VERSION = {v1},
}
Endnote :
%0 Unpublished work
%T Performance evaluation of the Mojette erasure code for fault-tolerant distributed hot data storage
%+ Institut de Recherche en Communications et en Cybernétique de Nantes (IRCCyN)
%+ Rozo Systems
%A Pertin, Dimitri
%A Féron, Didier
%A Van Kempen, Alexandre
%A Parrein, Benoît
%8 2015-05-27
%D 2015
%Z 1504.07038
%Z Computer Science [cs]/Information Theory [cs.IT]
%Z Computer Science [cs]/Networking and Internet Architecture [cs.NI]Preprints, Working Papers, ...
%X Packet erasure codes are today a real alternative to repli-cation in fault tolerant distributed storage systems. In this paper, we propose the Mojette erasure code based on the Mojette transform, a formerly tomographic tool. The performance of coding and decoding are compared to the Reed-Solomon code implementations of the two open-source reference libraries namely ISA-L and Jera-sure 2.0. Results clearly show better performances for our discrete geometric code compared to the classical algebraic approaches. A gain factor up to 2 is measured in comparison with the ISA-L Intel. Those very good performances allow to deploy Mojette erasure code for hot data distributed storage and I/O intensive applications.
%G English
%2 https://hal.archives-ouvertes.fr/hal-01156443/document
%2 https://hal.archives-ouvertes.fr/hal-01156443/file/hotstorage15-paper25.pdf
%L hal-01156443
%U https://hal.archives-ouvertes.fr/hal-01156443
%~ CNRS
%~ EC-NANTES
%~ IRCCYN
%~ UNAM
%~ UNIV-NANTES
%~ IRCCYN-SLP
Communication dans un congrès
Parrein Benoît, Lacan Jérôme, Normand Nicolas, Pertin Dimitri, Detchart Jonathan, Van Kempen Alexandre
FEC4Cloud: a research project promoting erasure coding for Cloud storage architectures
Rendez-vous de la Recherche et de l'Enseignement de la Sécurité des Systèmes d'Information (RESSI), May 2015, Troyes, France
Bibtext :
@inproceedings{parrein:hal-01156462,
TITLE = {{FEC4Cloud: a research project promoting erasure coding for Cloud storage architectures}},
AUTHOR = {Parrein, Beno{\^i}t and Lacan, J{\'e}r{\^o}me and Normand, Nicolas and Pertin, Dimitri and Detchart, Jonathan and Van Kempen, Alexandre},
URL = {https://hal.archives-ouvertes.fr/hal-01156462},
BOOKTITLE = {{Rendez-vous de la Recherche et de l'Enseignement de la S{\'e}curit{\'e} des Syst{\`e}mes d'Information (RESSI)}},
ADDRESS = {Troyes, France},
YEAR = {2015},
MONTH = May,
KEYWORDS = {RozoFS ; erasure coding ; Reed-Solomon codes ; Mojette transform ; discrete geometry ; cloud storage},
HAL_ID = {hal-01156462},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Paper
%F Oral
%T FEC4Cloud: a research project promoting erasure coding for Cloud storage architectures
%+ Institut de Recherche en Communications et en Cybernétique de Nantes (IRCCyN)
%+ Département de Mathématiques, Informatique, Automatique (DMIA)
%+ Rozo Systems
%A Parrein, Benoît
%A Lacan, Jérôme
%A Normand, Nicolas
%A Pertin, Dimitri
%A Detchart, Jonathan
%A Van Kempen, Alexandre
%< avec comité de lecture
%B Rendez-vous de la Recherche et de l'Enseignement de la Sécurité des Systèmes d'Information (RESSI)
%C Troyes, France
%8 2015-05
%D 2015
%K RozoFS
%K erasure coding
%K Reed-Solomon codes
%K Mojette transform
%K discrete geometry
%K cloud storage
%Z Computer Science [cs]/Cryptography and Security [cs.CR]
%Z Computer Science [cs]/Information Theory [cs.IT]
%Z Computer Science [cs]/Signal and Image ProcessingConference papers
%X The aim of FEC4Cloud project is to promote erasure codes in fault-tolerant Cloud storage (and computing) architectures. For this purpose, replication is commonly used to maintain the availability rate within the system. But, replication i.e triplication is storage capacity consuming. A network area of storage need usually two times more space of storage with replication than with erasure code approaches for the same level of protection (no more than four erasures basically). Intuitively, the complexity of erasure code could be opposed to real time read/write of I/O intensive application (with hot data). That leads to erasure code for archiving application traditionally (with some cold data). In this talk, we want to demonstrate (practically) that some erasure code can be used both for cold and hot data accesses. In FEC4Cloud, we investigate, in parallel, algebraic approach like the Reed-Solomon codes and discrete geometric approach supported notably by the Mojette transform (a discrete and exact Radon transform that will be presented). At this time, the implementations based on Mojette transform provide the best throughput (in MB/s) in encoding and decoding but it is possible (in very particular configurations) to accelerate highly the classical algebraic approach. In this talk, we will provide a comparison with coding schemes of [1] and [4] with a last update of results presented in [2] that brings a huge advantage to discrete geometric approaches (up to 6 time faster of reference Reed-Solomon coding schemes recently optimized in ISA library of Intel). The talk will also include results on several experimentations of a Mojette erasure-coded distributed file system (called RozoFS). In particular, we will focus on a real testbed that provides video editing on line service based on RozoFS to show that it is possible to use erasure code not only for archiving but also for I/O intensive applications [3]. Last results on privacy in cloud storage architecture will conclude the talk.
%G English
%L hal-01156462
%U https://hal.archives-ouvertes.fr/hal-01156462
%~ UNIV-NANTES
%~ IRCCYN
%~ IRCCYN-IVC
%~ CNRS
%~ EC-NANTES
%~ UNAM
%~ IVC-IPI
Communication dans un congrès
Pertin Dimitri, Normand Nicolas
Re-projection without Reconstruction
9ème Journées du Groupe de travail de Géométrie Discrète, Reims Image 2014, Nov 2014, Reims, France. pp.43
Bibtext :
@inproceedings{pertin:hal-01164872,
TITLE = {{Re-projection without Reconstruction}},
AUTHOR = {Pertin, Dimitri and Normand, Nicolas},
URL = {https://hal.archives-ouvertes.fr/hal-01164872},
BOOKTITLE = {{9{\`e}me Journ{\'e}es du Groupe de travail de G{\'e}om{\'e}trie Discr{\`e}te, Reims Image 2014}},
ADDRESS = {Reims, France},
PAGES = {43},
YEAR = {2014},
MONTH = Nov,
KEYWORDS = {discrete tomography ; redundancy ; convolution ; storage ; ghost},
PDF = {https://hal.archives-ouvertes.fr/hal-01164872/file/ri2014.pdf},
HAL_ID = {hal-01164872},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Proceedings
%T Re-projection without Reconstruction
%+ Institut de Recherche en Communications et en Cybernétique de Nantes (IRCCyN)
%+ Rozo Systems
%A Pertin, Dimitri
%A Normand, Nicolas
%< avec comité de lecture
%B 9ème Journées du Groupe de travail de Géométrie Discrète, Reims Image 2014
%C Reims, France
%P 43
%8 2014-11-25
%D 2014
%K discrete tomography
%K redundancy
%K convolution
%K storage
%K ghost
%Z Computer Science [cs]
%Z Mathematics [math]/Information Theory [math.IT]
%Z Computer Science [cs]/Computational Geometry [cs.CG]Conference papers
%X Discrete tomography focuses on image representation by its discrete projections, and the related inversion algorithms (or image reconstruction). Our study is based on redundant representations (considering more than just few projections). We propose a new approach to compute further redundancy (i.e. new projections) from a set of existing projections. While this technique relies on the geometric properties of ghosts, which are elements of the 2D image that sum to zero following some projection directions, we show an equivalent method using 1D convolutions, thus avoiding the explicit image reconstruction. This technique has interesting applications in distributed storage systems, where the use of redundancy data is key for system reliability.
%G English
%2 https://hal.archives-ouvertes.fr/hal-01164872/document
%2 https://hal.archives-ouvertes.fr/hal-01164872/file/ri2014.pdf
%L hal-01164872
%U https://hal.archives-ouvertes.fr/hal-01164872
%~ UNIV-NANTES
%~ CNRS
%~ EC-NANTES
%~ IRCCYN
%~ IRCCYN-IVC
%~ INSMI
%~ UNAM
Communication dans un congrès
Parrein Benoît, Pertin Dimitri, Normand Nicolas, Didier Féron
RozoFS: a fault tolerant I/O intensive distributed file system based on Mojette erasure code
Workshop Autonomic, Oct 2014, Toulouse, France
Bibtext :
@inproceedings{parrein:hal-01150741,
TITLE = {{RozoFS: a fault tolerant I/O intensive distributed file system based on Mojette erasure code}},
AUTHOR = {Parrein, Beno{\^i}t and Pertin, Dimitri and Normand, Nicolas and Didier, F{\'e}ron},
URL = {https://hal.archives-ouvertes.fr/hal-01150741},
BOOKTITLE = {{Workshop Autonomic}},
ADDRESS = {Toulouse, France},
YEAR = {2014},
MONTH = Oct,
HAL_ID = {hal-01150741},
HAL_VERSION = {v1},
}
Endnote :
%0 Conference Paper
%F Oral
%T RozoFS: a fault tolerant I/O intensive distributed file system based on Mojette erasure code
%+ Institut de Recherche en Communications et en Cybernétique de Nantes (IRCCyN)
%+ Rozo Systems
%A Parrein, Benoît
%A Pertin, Dimitri
%A Normand, Nicolas
%A Didier, Féron
%< avec comité de lecture
%B Workshop Autonomic
%C Toulouse, France
%8 2014-10-16
%D 2014
%Z Computer Science [cs]/Networking and Internet Architecture [cs.NI]Conference papers
%G English
%L hal-01150741
%U https://hal.archives-ouvertes.fr/hal-01150741
%~ UNIV-NANTES
%~ IRCCYN
%~ IRCCYN-IVC
%~ CNRS
%~ EC-NANTES
%~ UNAM
Voir toutes les publications