Télécom SudParis Djamal Zeghlache Professor Département Réseaux et Services Multimédia Mobiles Resource Management Group (in wireless, fixed and computer networks)
Département RS2M Méthodes, modèles et optimisation Optimisation Combinatoire, Apprentissage et Traitement Modèles Distribué de l’Information Méthodes Test et Validation de Optimisation Protocoles et de Services Apprentissage Prédiction Analyse Monitoring Vérification de Propriété, Services Gestion de ressources, Protocoles programmabilité et configuration des réseaux Architecture de Services et Réseaux Mobiles Architecture Services M ultimédia M obiles & Gestion de Ressources Réseaux Emergents IoT (IoT, capteurs, Ad Hoc, LTE, SF) Architectures e de Réseaux et de Services
Equipe (personnes) concernée (s) Cadre Gestion de ressources, programmabilité et configuration 2 E/ C, 6 doctorants, 1 Post Doc, 1 CIFRE Collaborations avec d’autres E/ C (RS2M , RST, INF et UM R CNRS 5157/ Samovar Journée Cloud sur Nantes Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Inter-cloud networking framework Work by Marouen Mechtri - Doctoral Student Journée Cloud sur Nantes 4 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Cloud Broker Framework 1 2 4 3 3 5 5 6 VLAN VLAN Journée Cloud sur Nantes Journée Cloud sur Nantes 5 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes. 18-19 Septembre 2014, Mines Nantes.
Cloud Networking Gateway Manager M. Mechtri - Doctorant The CNG Manager has: Northbound interface towards • client requesting connectivity based on the OCCI specification and service model. The CNG Manager Core selects • the appropriate drivers, in line with user expressed networking requirements. Southbound interface interacting • with transport technologies through specific drivers. Journée Cloud sur Nantes 6 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Cloud Networking Gateway and Gateway Manager implementations (M. Mechtri) Code available with readme at https://github.com/MarouenMechtri/CNG-Manager o You can also find links to the following (Mechtri et Ghribi) Installation of Openstack IceHouse release with Flat and o Neutron networking, Installation of Heat, o Deployment of Dockers o Journée Cloud sur Nantes 7 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Orchestration Users / Consumers / Tenants End users services and Applications Automatisation Service Requests (avec Dpt INF-TSP) Service and SLA descriptions and requests QoS & QoE (e.g. WS-Agreement and USDL) feedback Service Manager (reconfiguration decisions) Reconfiguration Autonomic Manager Alerts/notifications Manager (SLA injection & Mngt) SLA violations, {StrategySet} Service Life-Cycle Management degradations, Alerts/notifications anomalies SLA violations, Notifications/Message Exchange degradations, Monitoring Monitoring anomalies configuration configuration & Probe & Probe SLA & Performance Deployment SLA & Performance Deployment Monitoring of Analyzer Analyzer Monitoring of Service Instances MAPE {RuleSet} Service Instances {RuleSet} (+analysis) (analysis) Infrastructure Resources/services provisioning (Includes scheduling and smart placement) Cloud services Network services provisioning provisioning Managed Resources and Service Instances . . Cloud Provider . . Cloud Provider Network Providers (Heterogeneous PaaS & IaaS) (Heterogeneous PaaS & IaaS) Heterogeneous Networks & Networking technologies OCNI Cloud Extended OCNI Extended OCNI Interface Manager OCCI Interface OCCI Interface Interface Cloud Interface for IaaS & Paas Networking Controler VLAN or Networking SDN Controler Manager (Nox/Pox/SDN/Any other) Controler & Network API e.g. Software Networking technologies drivers Gateway in a VM Tunnel Driver NFV Driver OpenFlow NOX Other Drivers (e.g. OpenVPN, IPS EC, GRE … ) (e.g. Firewall,DNS,NAT… ) Driver Driver (e.g. hardware driver) VM i Probes/Monitor VM j Application flows Probe In-network services Probe PaaS - Containers PaaS - Containers e.g. OpenVSwitch Journée Cloud sur Nantes 8 e.g. Linux Bridge, e.g. OpenStack Neutron Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes. VLANs, VXLANs... Configurable substrate network OpenNebula.org (4.2)
Smart placement (Collaborations: M. Hadji à l’ISx depuis et doctorants de l’équipe) Journée Cloud sur Nantes 9 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Journée Cloud sur Nantes 10 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Energy based VM placement algorithm Objective, conditions & constraints Objective o initial VM placement leading to minimum number of used servers (or containers) Mathematical Programming Formulation modelled as a bin packing problem with a minimum power o consumption objective Variable comment m Number of servers Server power consumption limit P j,Max P j, current Current power consumption e j Boolean = 1 if j hosts VM x ij Boolean = 1 if VM I assigned to server j n Number of requested VMs Bound on the number of used servers Journée Cloud sur Nantes 11 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
VM placement algorithm Model variables Journée Cloud sur Nantes 12 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
VM Migration algorithm (Energy centric) Objective o Optimize data center power consumption using dynamic VM consolidation Mathematical Programming Formulation o Based on linear integer programming formulation • Z ijk = 1 if VM k migrated from server i to j • y i = 1 if server i idle and = 0 if at least one VM is active • m’ = number of non idle servers m’< m • P ’ k = power cost when migrating VM k • q i is the total number of VMs hosted on server i and candidate for migration into destination servers, especially server j in equation Journée Cloud sur Nantes 13 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
VM Migration algorithm Maximize number of empty servers to shut them down by migrating VM to achieve consolidation if a VMk is migrated from a server i (source) to a server j (destination), it can not be migrated to any other server l (l j). Ensuing migrations forbidden Destination VM power budget limit has to be respected Non idle servers candidate for migration have to be entirely emptied Equivalent total number of empty servers Do not migrate a VM whose job is about to end…. Journée Cloud sur Nantes 14 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
VM Migration algorithm A server candidate to a migration A VM can not be migrated to many should not migrate its own VMs servers at the same time Journée Cloud sur Nantes 15 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Exact energy efficient graph precoloring Algorithm Notations Variable comment Global graph of requests. G Vertex set of G. V The set of all edges in the graph G. E The set of intra VM edges. E’ C j,id A virtual resource unit VRU or color where j is the server to which it belongs and where id is its associated id. Cluster of colors containing colors that belong to the same server j. C j w j Performance per watt (PPW) of the server j. Z c Boolean = 1 if color c is used and 0 otherwise. x u c Boolean = 1 if node u is reserved to color c and 0 otherwise. Boolean = 1 if at least one color belonging to C j is used and 0 otherwise. y j n Total number of nodes in the graph G. m Number of servers of the data center. Journée Cloud sur Nantes 16 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Exact energy efficient graph precoloring Algorithm Objective: Advanced resource reservation (with minimum number of resources) while maximizing energy efficiency. Mathematical Programming Formulation : Minimize the number of used resources while maximizing energy efficiency Ensure that each node is associated to one and only one color Any two nodes connected by an edge must have different colors Ensure that z c is equal to 1 if the color c is assigned to a node u Nodes belonging to the same VM have to be associated to colors of the same color cluster Journée Cloud sur Nantes 17 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Exact Journée Cloud sur Nantes 18 Resource Management Group (in wireless, fixed and computer networks) - D. Zeghlache 18-19 Septembre 2014, Mines Nantes.
Recommend
More recommend
