Abstract

Transport ecosystems that combine software defined networking (SDN) and network function virtualization (NFV) are characterized by an unprecedented network control and resource dynamicity. Manual optimization is unmanageable. In this context, open systems that manage and orchestrate SDN/NFV-enabled networks offer programming frameworks that abstract the low-level particularities in the data-plane forwarding devices and in the hardware appliances that provide the IT resources. Although these open systems present notable complexity, their programming abstractions promote a client layer where third-party applications can provide different functionalities thus enabling optimization-as-a-service business opportunities. In this paper, we cover open-source optimization software initiatives for offline planning and online provisioning and orchestration of SDN/NFV networks. With this goal in mind, we first focus on open software (and framework) initiatives through a set of realistic use cases that require optimization in multi-layer optical transport scenarios and ecosystems that combine transport with IT resources. The importance of a joint optimization of both network and IT domains is emphasized, a new paradigm triggered by SDN/NFV technologies. We discuss the theoretical limits to algorithm performances, and review available open-source frameworks for problem modeling that enable the interaction with solvers. Finally, we focus on the Net2Plan open-source network planning tool, a Java-based software that suitably embraces the multiple features required in the optimization of joint transport network and IT resource SDN/NFV ecosystems. Recent works based on Net2Plan are reviewed to illustrate its suitability for rapid algorithm prototyping, and for interaction with SDN/NFV-enabled networks.

Highlights

  • T ELECOM operators’ infrastructure is undergoing through major challenges due to the need to support the forecasted traffic growth which presents high levels of dynamicity and a wide variety of requirements [2]

  • Examples are CPU pinning option that assigns a vCPU to a particular host CPU, and the (EPA) templates that permit defining Non-Uniform Memory Access (NUMA) topologies in which system memory is divided into cells or nodes that are associated with particular CPUs [45]

  • The combination of software defined networking (SDN) and network function virtualization (NFV) is fostering transport ecosystems characterized by unprecedented network control and unique resource dynamicity

Read more

Summary

INTRODUCTION

T ELECOM operators’ infrastructure is undergoing through major challenges due to the need to support the forecasted traffic growth which presents high levels of dynamicity and a wide variety of requirements [2]. The fundamental difference between transport (SDN) [5], [9] and IT resources (NFV) [6], [8], [10] clearly implies a different set of features to be abstracted in each open system, which depends on network types, specific software architectures (e.g., control plane implementations) and management strategies Each of these open initiatives address specific needs ranging from common application programming interfaces (APIs) definition for control and management of network systems [12]–[15] to resource orchestration in multi-tenant infrastructures [10], [16].

USE CASES
Optical-Layer Transport Control
Control of Multi-Layer IP Over WDM Transport Networks
Joint Transport Network and IT Resource Allocation
ADDRESSING THE USE CASES
Theoretical Limits to Algorithm Performances
Practical Aspects in Solver Interactions
NETWORK OPTIMIZATION TOOLS
Vendor Agnostic WDM Line Engineering
IP Over WDM Optimization via SDN Controllers
NFV Optimization
Joint IT and Multi-Layer Network Resource Provisioning
Findings
SUMMARY

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.