Skip to main content

EVOLUCION HACIA REDES Y SERVICIOS AUTO-GESTIONADOS PARA EL 5G DEL FUTURO

PID2019-108713RB-C52

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Call Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Call year 2019
State Plan 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSIDAD CARLOS III DE MADRID

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

DQN-based intelligent controller for multiple edge domains

  • Llorens Carrodeguas, Alejandro|||0000-0002-4329-7962
  • Cervelló Pastor, Cristina|||0000-0002-8056-0774
  • Valera Pintor, Francisco
Advanced technologies like network function virtualization (NFV) and multi-access edge computing (MEC) have been used to build flexible, highly programmable, and autonomously manageable infrastructures close to the end-users, at the edge of the network. In this vein, the use of single-board computers (SBCs) in commodity clusters has gained attention to deploy virtual network functions (VNFs) due to their low cost, low energy consumption, and easy programmability. This paper deals with the problem of deploying VNFs in a multi-cluster system formed by this kind of node which is characterized by limited computational and battery capacities. Additionally, existing platforms to orchestrate and manage VNFs do not consider energy levels during their placement decisions, and therefore, they are not optimized for energy-constrained environments. In this regard, this study proposes an intelligent controller as a global allocation mechanism based on deep reinforcement learning (DRL), specifically on deep Q-network (DQN). The conceived mechanism optimizes energy consumption in SBCs by selecting the most suitable nodes across several clusters to deploy event requests in terms of nodes’ resources and events’ demands. A comparison with available allocation algorithms revealed that our solution required 28% fewer resource costs and reduced 35% the energy consumption in the clusters’ computing nodes while maintaining high levels of acceptance ratio., This work has been supported in part (50%) by the Agencia Estatal de Investigación of Ministerio de Ciencia e Innovación of Spain under projects PID2019-108713RB-C51 & PID2019-108713RB-C52 MCIN/ AEI/10.13039/501100011033; and in part (50%) by AI@EDGE H2020ICT-52-2020 under grant agreement No. 10101592.
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

DQN-based intelligent controller for multiple edge domains

  • Llorens Carrodeguas, Alejandro
  • Cervelló Pastor, Cristina
  • Valera Pintor, Francisco
Advanced technologies like network function virtualization (NFV) and multi-access edge computing (MEC) have been used to build flexible, highly programmable, and autonomously manageable infrastructures close to the end-users, at the edge of the network. In this vein, the use of single-board computers (SBCs) in commodity clusters has gained attention to deploy virtual network functions (VNFs) due to their low cost, low energy consumption, and easy programmability. This paper deals with the problem of deploying VNFs in a multi-cluster system formed by this kind of node which is characterized by limited computational and battery capacities. Additionally, existing platforms to orchestrate and manage VNFs do not consider energy levels during their placement decisions, and therefore, they are not optimized for energy-constrained environments. In this regard, this study proposes an intelligent controller as a global allocation mechanism based on deep reinforcement learning (DRL), specifically on deep Q-network (DQN). The conceived mechanism optimizes energy consumption in SBCs by selecting the most suitable nodes across several clusters to deploy event requests in terms of nodes’ resources and events’ demands. A comparison with available allocation algorithms revealed that our solution required 28% fewer resource costs and reduced 35% the energy consumption in the clusters’ computing nodes while maintaining high levels of acceptance ratio., This work has been supported in part (50%) by the Agencia Estatal de Investigación of Ministerio de Ciencia e Innovación of Spain under projects PID2019-108713RB-C51 & PID2019-108713RB-C52 MCIN/ AEI/10.13039/501100011033; and in part (50%) by AI@EDGE H2020ICT-52-2020 under grant agreement No. 10101592., Postprint (published version)
e_Buah Biblioteca Digital Universidad de Alcalá
Publication Artículo científico (JournalArticle)

Stateless flow-zone switching using software-defined addressing

  • González Diaz, Sergio
  • Marks, Roger
  • Rojas Sánchez, Elisa|||0000-0002-6385-2628
  • Oliva, Antonio de la
  • Gazda, Robert
The trend toward cloudification of communication networks and services, with user data and applications stored and processed in data centers, pushes the limits of current Data Center Networks (DCNs), requiring improved scalability, resiliency, and performance. Here we consider a DCN forwarding approach based on software-defined addressing (SDA), which embeds semantics in the Medium Access Control (MAC) address and thereby enables new forwarding processes. This work presents Flow-Zone Switching (FZS), a loop-free location-based source-routing solution that eliminates the need for forwarding tables by embedding routing instructions and flow identifiers directly in the flow-zone software-defined address. FZS speeds the forwarding process, increasing the throughput and reducing the latency of QoS-sensitive flows while reducing the capital and operational costs of switching. This paper presents details of FZS and a performance evaluation within a complete DCN., European Commission, Agencia Estatal de Investigación, Comunidad de Madrid