Survey on Network Virtualization Hypervisors for Software Defined Networking

Abstract: Software defined networking (SDN) has emerged as a promising paradigm for making the control of communication networks flexible. SDN separates the data packet forwarding plane, i.e., the data plane, from the control plane and employs a central controller. Network virtualization allows the flexible sharing of physical networking resources by multiple users (tenants). Each tenant runs its own applications over its virtual network, i.e., its slice of the actual physical network. The virtualization of SDN networks promises to allow networks to leverage the combined benefits of SDN networking and network virtualization and has therefore attracted significant research attention in recent years. A critical component for virtualizing SDN networks is an SDN hypervisor that abstracts the underlying physical SDN network into multiple logically isolated virtual SDN networks (vSDNs), each with its own controller. We comprehensively survey hypervisors for SDN networks in this article. We categorize the SDN hypervisors according to their architecture into centralized and distributed hypervisors. We furthermore sub-classify the hypervisors according to their execution platform into hypervisors running exclusively on general-purpose compute platforms, or on a combination of general-purpose compute platforms with general- or special-purpose network elements. We exhaustively compare the network attribute abstraction and isolation features of the existing SDN hypervisors. As part of the future research agenda, we outline the development of a performance evaluation framework for SDN hypervisors.

• The paper systematically categorizes SDN hypervisors by comparing centralized models like FlowVisor with distributed architectures such as FlowN to address scalability and isolation challenges.
• It introduces a dual-step benchmarking methodology that evaluates both integrated hypervisor performance and individual virtualization functions.
• Future work is suggested to enhance SLA definitions, develop comprehensive benchmarks, and design scalable hypervisors for large-scale network deployments.

A Comprehensive Survey of Network Virtualization Hypervisors for Software Defined Networking

The evolution of networking paradigms has witnessed a significant transformation with the advent of Software Defined Networking (SDN) and network virtualization. SDN, characterized by its separation of the control plane from the data plane, presents a flexible and programmable approach to managing networking infrastructures. Network virtualization builds upon this paradigm, enabling multiple tenants to share the same physical infrastructure through logically isolated virtual networks, also known as network slices. The survey paper by Andreas Blenk et al. meticulously categorizes, reviews, and evaluates hypervisors, the key components for realizing network virtualization in SDN environments.

Key Contributions

The paper systematically categorizes SDN hypervisors based on their architectural and operational characteristics. Two primary architectural classifications are identified: centralized and distributed hypervisors. Centralized hypervisors operate as a single logical entity managing multiple switches, while distributed hypervisors spread the virtualization functions across multiple points in the network to enhance scalability and resilience.

Detailing the Landscape of SDN Hypervisors

The study categorizes existing centralized hypervisors into those designed for general networks, special network types (such as optical or wireless networks), and policy-based hypervisors aimed at managing heterogeneous SDN control applications. FlowVisor, recognized as the pioneering SDN hypervisor, serves as a baseline for comparison. FlowVisor operates as an intermediary between physical network resources and tenant controllers, maintaining isolation through mechanisms dealing with flowspace definition and rule-based interactions.

Distributed Hypervisor Architectures

In response to concerns over the scalability of centralized models, distributed architectures such as FlowN and Network Hypervisor emerge. These solutions partition the physical infrastructure into distinct administrative domains. For instance, AutoSlice and OpenVirteX provide mechanisms to separately manage and control these domains, facilitating seamless and efficient operations across larger network topologies.

Evaluation Metrics for SDN Hypervisors

Blenk et al. underscore the necessity for a robust performance evaluation framework for SDN hypervisors. They propose a dual-step benchmarking method—evaluating the hypervisor as an integrated system, followed by a detailed assessment of its individual virtualization functions. This approach is vital for understanding not only the overhead introduced by the hypervisor but also its impact on network performance and the effective utilization of network resources.

Implications and Future Directions

The survey posits that SDN hypervisors are essential for effective network resource management, enabling scalable and flexible network service deployments. Looking forward, the paper suggests further research into areas such as service level agreement definition for vSDNs, the development of comprehensive hypervisor benchmarks, and scalable designs for hypervisor architectures to meet the pragmatic constraints of large-scale deployments.

Conclusion

The comprehensive survey by Andreas Blenk and collaborators provides critical insights into the current state and future trajectory of SDN hypervisors. While significant progress has been made, the paper identifies key areas that require further investigation to foster the evolution of network virtualization in tandem with SDN advancements. The balance between innovation in abstraction and the practical demands of efficient and reliable network operations remains a pivotal challenge in the continued development of this domain.