TY - GEN
T1 - Deep Reinforcement Learning for Topology-Aware VNF Resource Prediction in NFV Environments
AU - Jalodia, Nikita
AU - Henna, Shagufta
AU - Davy, Alan
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/11
Y1 - 2019/11
N2 - Network Function Virtualisation (NFV) has emerged as a key paradigm in network softwarisation, enabling virtualisation in future generation networks. Once deployed, the Virtual Network Functions (VNFs) in an NFV application's Service Function Chain (SFC) experience dynamic fluctuations in network traffic and requests, which necessitates dynamic scaling of resource instances. Dynamic resource management is a critical challenge in virtualised environments, specifically while balancing the trade-off between efficiency and reliability. Since provisioning of virtual infrastructures is time-consuming, this negates the Quality of Service (QoS) requirements and reliability criterion in latency-critical applications such as autonomous driving. This calls for predictive scaling decisions to balance the provisioning time sink, with a methodology that preserves the topological dependencies between the nodes in an SFC for effective resource forecasting. To address this, we propose the model for an Asynchronous Deep Reinforcement Learning (DRL) enhanced Graph Neural Networks (GNN) for topology-aware VNF resource prediction in dynamic NFV environments.
AB - Network Function Virtualisation (NFV) has emerged as a key paradigm in network softwarisation, enabling virtualisation in future generation networks. Once deployed, the Virtual Network Functions (VNFs) in an NFV application's Service Function Chain (SFC) experience dynamic fluctuations in network traffic and requests, which necessitates dynamic scaling of resource instances. Dynamic resource management is a critical challenge in virtualised environments, specifically while balancing the trade-off between efficiency and reliability. Since provisioning of virtual infrastructures is time-consuming, this negates the Quality of Service (QoS) requirements and reliability criterion in latency-critical applications such as autonomous driving. This calls for predictive scaling decisions to balance the provisioning time sink, with a methodology that preserves the topological dependencies between the nodes in an SFC for effective resource forecasting. To address this, we propose the model for an Asynchronous Deep Reinforcement Learning (DRL) enhanced Graph Neural Networks (GNN) for topology-aware VNF resource prediction in dynamic NFV environments.
KW - Asynchronous Deep Q-Learning
KW - Deep Learning
KW - Deep Reinforcement Learning
KW - Dynamic Resource Prediction
KW - Future Generation Networks
KW - Graph Neural Networks
KW - Machine Learning
KW - NFV
KW - Prediction
KW - Topology Awareness
UR - http://www.scopus.com/inward/record.url?scp=85082990963&partnerID=8YFLogxK
U2 - 10.1109/NFV-SDN47374.2019.9040154
DO - 10.1109/NFV-SDN47374.2019.9040154
M3 - Conference contribution
AN - SCOPUS:85082990963
T3 - IEEE Conference on Network Function Virtualization and Software Defined Networks, NFV-SDN 2019 - Proceedings
BT - IEEE Conference on Network Function Virtualization and Software Defined Networks, NFV-SDN 2019 - Proceedings
A2 - Horner, Larry
A2 - Tutschku, Kurt
A2 - Granelli, Fabrizio
A2 - Sekiya, Yuji
A2 - Tacca, Marco
A2 - Bhamare, Deval
A2 - Parzyjegla, Helge
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2019 IEEE Conference on Network Function Virtualization and Software Defined Networks, NFV-SDN 2019
Y2 - 12 November 2019 through 14 November 2019
ER -