Ali Hammad

Transport Network Orchestration for End-to-End Multilayer Provisioning Across Heterogeneous SDN/OpenFlow and GMPLS/PCE Control Domains

A multidomain optical transport network composed of heterogeneous optical transport technologies (e.g., flexi/fixed-grid optical circuit switching and optical packet switching) and control plane technologies (e.g., centralized OpenFlow or distributed GMPLS) does not naturally interoperate, and a network orchestration mechanism is required. A network orchestrator allows the composition of end-to-end network service provisioning across multidomain optical networks comprising different transport and control plane technologies. Software-defined networking (SDN) is a key technology to address this requirement, since the separation of control and data planes makes the SDN a suitable candidate for end-to-end provisioning service orchestration across multiple domains with heterogeneous control and transport technologies. This paper presents two different network orchestrations architectures based on the application-based network operations (ABNO) which is being defined by IETF based on standard building blocks. Then, we experimentally assesses in the international testbed of the STRAUSS project, an ABNO-based network orchestrator for end-to-end multi-layer (OPS and Flexi-grid OCS) and multidomain provisioning across heterogeneous control domains (SDN/OpenFlow and GMPLS/Stateful PCE) employing dynamic domain abstraction based on virtual node aggregation.

Experimental assessment of ABNO-based network orchestration of end-to-end multi-layer (OPS/OCS) provisioning across SDN/OpenFlow and GMPLS/PCE control domains

We present and experimentally assess in an international testbed an ABNO-based network orchestrator for end-to-end multi-layer (OPS and Flexi-grid OCS) and multi-domain provisioning across heterogeneous control domains (SDN/OpenFlow and GMPLS/Stateful PCE) employing dynamic domain abstraction based on virtual node aggregation.

Network virtualization over elastic optical networks with different protection schemes

With network virtualization, the physical infrastructure can be partitioned into multiple parallel virtual networks for sharing purposes. However, different transport technologies or quality of service (QoS) levels may impact both the requested amount of resources and the characteristics of different virtual instances that can be built on top of a single physical infrastructure. In this paper we propose a novel mixed integer linear programming (MILP) formulation for different schemes of protection in scenarios where multiple virtual topologies run over an elastic optical network. The proposed MILP formulation uses the concept of bandwidth squeezing to guarantee a minimum bandwidth for surviving virtual topologies. It achieves a high level of survivability for traffic that is subject to a different committed service profile for each virtual topology. Case studies are carried out in order to analyze the basic properties of the formulation in small networks, and three heuristics are proposed for larger networks.

Demonstration of Virtualizeable and Software-Defined Optical Transceiver

In order to serve the future high-performance network-based Internet applications, optical network virtualization is proposed to offer each application type a dedicated virtual optical network (VON). Virtualizeable bandwidth variable transceiver (V-BVT) is a key enabler in supporting the creation of multiple VONs. In this paper, we present a feasible V-BVT architecture that can be a part of a software-defined optical network. The proposed V-BVT has a novelty to offer independent operation, control, and management abilities to the clients or higher level network controllers. In addition, a specific V-BVT virtualization algorithm is proposed, in order to enable the efficient creation of multiple coexisting, but independent virtual transceivers that share the same V-BVT physical resources. The virtual transceiver can provide specific bit rate, subcarrier, modulation format, and a corresponding baud rate to each VON, based on the requirement of the VON demand, V-BVT resources availability, and optical network status. We further realize the proposed V-VBT architecture on an experimental platform with a software-defined network controller. The V-BVT resource allocation through the proposed virtualization algorithm is also performed using the extended OpenFlow protocol. The proposed and experimentally demonstrated V-BVT achieves independence in virtual transceivers control and management in the control plane, while maintaining the coexisting and isolation features in the physical layer.

Online and offline virtualization of optical transceiver

Future high-performance network-based applications are delivered over the high-capacity dynamic optical network. Each of these applications requires dedicated network service, which can be provided by the virtual optical network (VON) created by optical network virtualization. The virtualizable bandwidth variable transceiver (V-BVT) is a key enabling technology for optical network virtualization. In this paper, we propose the virtualization of V-BVT to support the virtualization of optical orthogonal frequency-division-multiplexing-based elastic optical network. We present a novel V-BVT architecture, and introduce both online and offline virtualization algorithms for V-BVT. Accordingly, multiple independent but coexisting virtual transceivers that share the same physical transceiver resources are created, in order to serve separate VONs. To guarantee the isolation of the created virtual transceivers together with their quality of transmission (QoT), the impact of physical layer impairments on different V-BVT solutions is also considered and integrated into the virtualization algorithms. In the offline virtualization, both heuristic and integral linear programming methods are proposed, in order to maximize the VON demand accommodation using the given physical V-BVT resources. In the online virtualization, a heuristic method is proposed to accommodate real-time received VON demands. By applying both algorithms, multiple virtual transceivers can be dynamically created based on the bandwidth and QoT of the VON demands. Finally, we evaluate and compare the performance of the proposed algorithms, and also verify the V-BVT transmission performance through simulation studies.

Cross-layer optimization of network resource virtualization in IP over O-OFDM networks

Optical orthogonal frequency division multiplexing (O-OFDM) is a promising transport technology for virtualization and bandwidth sharing over high-capacity optical fibers. In this paper, we propose an IP-over O-OFDM optical network architecture utilizing network virtualization across IP and optical layers. A design goal of this architecture is to provide dynamicity and flexibility in addition to the high capacity required to support multirate traffic with various quality of service requirements. In this architecture, optimized allocation of optical resources to virtual optical networks that accommodate the traffic of IP networks is considered. O-OFDM is utilized to allocate a set of low-rate contiguous subcarriers to each traffic connection represented by a virtual optical link. In addition, we study virtualization of IP network resources (i.e., routers and bandwidth capacities of links) in order to compose several customized virtual IP networks. Furthermore, coordination between IP and optical network virtualization mechanisms is investigated in order to achieve the required optimization of network resource virtualization across the different domains. For this purpose, a novel replanning approach is proposed and is aimed to adapt the resources allocated for virtual optical networks based on the requirements of an IP layer.

Approaches to maximize the open capacity of elastic optical networks

This paper proposes a linear formulation and an iterative heuristic, both with traffic grooming capability, which can maximize the number of remaining available routes and minimize the number of transceivers in Elastic Optical Networks (EON). The aim of the proposal is to preserve the open capacity for the accommodation of future unknown demands. Case studies are carried out in order to analyze the basic properties of the formulation in a small network, and the heuristic is used for moderate larger networks. The results suggest that it is feasible to preserve enough open capacity to avoid blocking of future requests in EON with scarce resources.

Impact of fog and cloud computing on an IoT service running over an optical/wireless network testbed

With the advance of the Internet of Things (IoT), the interaction between humans and smart objects is already a reality. New applications that are expected to operate in dynamic environments must support different modes of human/machine interaction (e.g., voice and sign language), exhibit same or better performance in heterogeneous wireless and optical networks, and be able to react in real time. In particular, dispersed computing has arisen as an approach to deal with latency issues in this context. In the work described herein, we design a smart lighting IoT system that allows control of light bulbs (turn on/off, color and brightness change) through voice and sign language. This work addresses the idea of dispersed computing, which is implemented through fog computing, and combines it with virtualized resources to mitigate latency in the convergence point between wireless and optical networks. The proof-of-concept implementation of our design demonstrates the viability of the approach.