The HTTP/2 protocol introduces the concept of request prioritization, where the client manifests to the server its preferences for completing important requests more quickly. In this work, we propose that this prioritization can also occur at the network layer with the help of a software-defined network. By classifying and marking the streams, which are related to the most critical resources needed for web page rendering, as high-priority traffic, it is possible to obtain a better Quality-of-Experience (QoE) for the web users. To demonstrate the benefits of the proposed approach, a prototype was built to emulate a software-defined wide area network, where paths with different Quality-of-Service (QoS) characteristics were available. The SpeedIndex metric was used to measure the QoE for the Alexa top 9 sites on the web. The obtained results indicate that the proposed approach can significantly improve the QoE in lossy environments, which are known to be challenging to the HTTP/2 protocol.

The management of multilayer networks is often a complex task, since each layer is typically independently operated and maintained, and there is a lack of shared network resource knowledge between layers. In this work, we have illustrated that Ant Colony Optimization (ACO) algorithms can be a suitable solution for coordinated management of disjoint IP/MPLS and optical layers with limited or no visibility between the control planes of each layer. ACO algorithms are used to achieve fully-distributed multilayer routing policies in IP/MPLS-over-optical networks. Simulations have demonstrated that ACO-based policies can achieve lower levels of blocking probability and higher levels of restorability for IP/MPLS connections compared to fixed-alternate policies in almost all cases.

Crankback re-routing extensions can offer significant improvements in the successful setup of Label Switched Paths (LSPs) by allowing new retries on alternate paths that circumvent blocked links or nodes. These extensions can be incorporated into a fully-distributed algorithm based on Ant Colony Optimization metaheuristics, taking advantage of its self-adapting, emergent behavior. By comparing with traditional fixed-alternate re-routing mechanisms, simulations have demonstrated that the proposed algorithm can efficiently mitigate lightpath blocking, both during normal operation and in case of network failure, by locally repairing failed LSP setups due to blocked/failed resources.

Optical networking paradigms based on statistical multiplexing, such as Optical Burst Switching or Optical Packet Switching, require the adoption of suitable contention resolution mechanisms at the optical nodes due to packets/bursts attempting to get access to the shared output channel at the same time. In the most general case of multi-fiber output links, contentions are tried to be solved by exploiting different domains — namely space, wavelength and time — and by applying an optimal scheduling policy. This paper focuses on the key parameters that can be used to design and optimize an optical packet/burst contention resolution scheme and shows how such parameters should be correlated for a correct performance assessment, taking into account feasibility trade-offs due to limited optical buffering and wavelength conversion capability.

In this paper, we propose an ACO-based algorithm that can adaptively balance the network load to mitigate congestion, and hence reduce the overall loss of packets in an OPS network. In addition, the proposed algorithm is fault-tolerant, with the recovery from a failed link or node starting even before detection and localization of the failure event. Since both routing and restoration rely on pheromone levels laid by ant-like agents, any failure will be promptly perceived by some ants, which will then try to circumvent the affected part of the network, thus adapting the routing to the new topology.

The use of a manageable optical network is an important requirement for the new advanced data-intensive grid applications that begin to emerge on the e-Science field. In this paper, we propose an ACO-based algorithm that can provision on-demand and dynamically lightpaths on a grid system. Indeed, the proposed algorithm can schedule jobs by discovering processing and network resources on the grid, assigning the job to a specific system and executing the job. This is achieved with the help of an integrated GMPLS control plane, which allows both resource monitoring and management of networking resources.

This paper proposes a methodology to compare scheduling algorithms for congestion resolution in optical burst switching or optical packet switching networks, focusing on the issue of wavelength conversion capabilities. The performance of such algorithms when full wavelength conversion is available at the switching nodes are compared with the case of conversion capabilities over a limited range of wavelengths. The results presented show that limited range wavelength conversion is not necessarily performing worse than full wavelength conversion, as long as the correct terms of comparison are considered.