Dynamic Resource Management in Service-Oriented Core Networks

This book provides a comprehensive study of dynamic resource management for network slicing in service-oriented fifth-generation (5G) and beyond core networks. It focuses on developing efficient computation resource provisioning and scheduling solutions to guarantee consistent service performance in terms of end-to-end (E2E) data delivery delay. The book utilizes queueing theory and techniques such as optimization and machine learning to address the challenges of dynamic traffic dynamics. It presents an optimization model for flow migration and a machine learning approach for dynamic VNF resource scaling and migration. The book is valuable for researchers, graduate students, and professionals in the industry seeking solutions to dynamic resource management in 5G and beyond networks.

Format: Paperback / softback
Length: 173 pages
Publication date: 05 November 2022
Publisher: Springer Nature Switzerland AG

This book offers a timely and comprehensive exploration of dynamic resource management for network slicing in service-oriented fifth-generation (5G) and beyond core networks. It delves into the development of efficient computation resource provisioning and scheduling solutions to ensure consistent service performance in terms of end-to-end (E2E) data delivery delay.

Network slicing is made possible by the software defined networking (SDN) and network function virtualization (NFV) paradigms. For a network slice with a specific target traffic load, virtual network function (VNF) placement and traffic routing with static resource allocations enable the E2E service delivery. However, when data traffic enters the network, the traffic load is dynamic and can deviate from the target value, potentially leading to Quality of Service (QoS) performance degradation and network congestion.

Data traffic exhibits dynamics at various time granularities. For instance, traffic statistics, such as mean and variance, can be non-stationary and experience significant changes in a coarse time granularity, which are often predictable. Within a long time duration with stationary traffic statistics, there are traffic dynamics in small timescales, which are typically highly bursty and unpredictable. To ensure continuous QoS performance guarantee and efficient and fair operation of network slices over time, it is essential to develop dynamic resource management schemes for the embedded services in the presence of traffic dynamics during virtual network operation.

Queueing theory serves as a framework for system modeling, and various techniques, including optimization and machine learning, are employed to address dynamic resource management challenges. Based on a simplified M/M/1 queueing model with Poisson traffic arrivals, an analytical solution is derived to optimize the resource allocation for network slicing. This solution demonstrates the trade-off between the utilization of resources and the delay experienced by the end users.

Furthermore, the book explores the application of deep learning in dynamic resource management for network slicing. It discusses the use of convolutional neural networks (CNNs) to predict traffic patterns and optimize resource allocation in real-time. The authors demonstrate how deep learning can be used to enhance the performance of network slicing by learning the characteristics of different traffic types and adapting resource allocation accordingly.

In conclusion, this book provides a valuable resource for researchers, practitioners, and policymakers interested in the field

This book offers a timely and comprehensive exploration of dynamic resource management for network slicing in service-oriented fifth-generation (5G) and beyond core networks. It delves into the development of efficient computation resource provisioning and scheduling solutions to ensure consistent service performance in terms of end-to-end (E2E) data delivery delay.

Network slicing is made possible by the software defined networking (SDN) and network function virtualization (NFV) paradigms. For a network slice with a specific target traffic load, virtual network function (VNF) placement and traffic routing with static resource allocations enable the E2E service delivery. However, when data traffic enters the network, the traffic load is dynamic and can deviate from the target value, potentially leading to Quality of Service (QoS) performance degradation and network congestion.

Data traffic exhibits dynamics at various time granularities. For instance, traffic statistics, such as mean and variance, can be non-stationary and experience significant changes in a coarse time granularity, which are often predictable. Within a long time duration with stationary traffic statistics, there are traffic dynamics in small timescales, which are typically highly bursty and unpredictable. To ensure continuous QoS performance guarantee and efficient and fair operation of network slices over time, it is essential to develop dynamic resource management schemes for the embedded services in the presence of traffic dynamics during virtual network operation.

Queueing theory serves as a framework for system modeling, and various techniques, including optimization and machine learning, are employed to address dynamic resource management challenges. Based on a simplified M/M/1 queueing model with Poisson traffic arrivals, an analytical solution is derived to optimize the resource allocation for network slicing. This solution demonstrates the trade-off between the utilization of resources and the delay experienced by the end users.

Furthermore, the book explores the application of deep learning in dynamic resource management for network slicing. It discusses the use of convolutional neural networks (CNNs) to predict traffic patterns and optimize resource allocation in real-time. The authors demonstrate how deep learning can be used to enhance the performance of network slicing by learning the characteristics of different traffic types and adapting resource allocation accordingly.

In conclusion, this book provides a valuable resource for researchers, practitioners, and policymakers interested in the

Weight: 296g
Dimension: 235 x 155 (mm)
ISBN-13: 9783030871383
Edition number: 1st ed. 2021