Theofanis Orphanoudakis

Efficient medium arbitration of FSAN-compliant GPONs

SUMMARY The steadily rising demand for multimedia and data services, the falling cost and omnipresence of Ethernet and the maturity of passive optical networks (PON) technology, promise to radically change the landscape in the local loop. The heart of a gigabit PON system (recently standardized by FSAN/ITU) is the medium access controller (MAC), which arbitrates access to the upstream link among users with fluctuating traffic demands and effects the multiplexing and concentration policy. At the same time, it has to safeguard the service quality and enforce the parameters agreed in the service level agreements (SLAs) between the users and the service provider. In this paper, a MAC protocol designed to serve any mix of services according to their quality of service (QoS) needs, employing four priority levels along with a high number of logically separate data queues is presented. The architecture and implementation in hardware of a MAC algorithm capable of allocating bandwidth down to a resolution of a byte with QoS differentiation is the focus of this paper. It employs the bandwidth arbitration tools of the FSAN/ITU G.984.3 standard and maps SLA parameters to GPON service parameters to create an efficient, fair and flexible residential access system. Copyright # 2005 John Wiley & Sons, Ltd.

Exploiting PONs for mobile backhaul

The growing popularity of mobile data services necessitates a rapid rise in network capacity not only on the air interface to the end user, but also in the backhaul network. The latter is quite important for the mobile operator business model, affecting capital investment, operational expenses, service deployment, and customer experience. Fiber infrastructure is inevitably the only long-term solution, and the deployment of passive optical networks presents an opportunity for a cost-effective, scalable, and future-proof solution. In this article we investigate the use of PONs for mobile backhaul and propose a resource allocation framework building on the efficiency of PONs to share resources, dynamically allocate bandwidth in real time, and enhance efficiency by improved statistical multiplexing. The main objective of this work is to exploit existing standardized technologies, and provide design and deployment guidelines regarding PON MAC operation, enabling a gradual and future-safe infrastructure upgrade of mobile backhaul systems.

An Optical Network Architecture With Distributed Switching Inside Node Clusters Features Improved Loss, Efficiency, and Cost

The novel core network architecture presented in this paper realizes distributed all-optical switching of payload by partitioning the network into a number of geographically limited domains, where two-way reservations are effective. Thus, inside each domain, loss is eliminated, while traffic from many nodes can be aggregated into single bursts, improving efficiency. Clustered nodes contribute contiguous optical slots, which are marshaled into composite optical frames destined for other clusters, under the guidance of a reservation-based control protocol. The lossless aggregation of traffic from several core nodes allows the use of cost-effective bufferless all-optical transport among the domains with electrical buffers employed at the periphery of the system. The end result is a triple improvement in loss probabilities, efficiency, and cost. This is achieved by exploiting three features of the architecture: the distributed switching functionality (as in early LANs when centralized switching was expensive), localized reservations (avoiding the intolerable delays of end-to-end reservations), and a reduced number of source-destination pairs (by means of node clustering into reservation domains)

Queue Management in Network Processors

One of the main bottlenecks when designing a network processing system is very often its memory subsystem. This is mainly due to the state-of-the-art network links operating at very high speeds and to the fact that in order to support advanced quality of service (QoS), a large number of independent queues is desirable. In this paper we analyze the performance bottlenecks of various data memory managers integrated in typical network processing units (NPU). We expose the performance limitations of software implementations utilizing the RISC processing cores typically found in most NPU architectures and we identify the requirements for hardware assisted memory management in order to achieve wire-speed operation at gigabit per second rates. Furthermore, we describe the architecture and performance of a hardware memory manager that fulfills those requirements. This memory manager, although it is implemented in a reconfigurable technology, can provide up to 6.2 Gbit/s of aggregate throughput, while handling 32 K independent queues.

Optical packet routers: how they can efficiently and cost-effectively scale to petabits per second

Todays data-centric networks are becoming progressively dynamic with respect to the transported traffic volume, to the spatial and temporal variations of traffic patterns, and to the subsequent interconnection request patterns. Optical packet routers (OPRs) aim to provide a viable answer to these requirements by rendering the optical layer adaptable, reconfigurable at will, and cost-effective by means of statistical multiplexing of the network resources while satisfying end-to-end quality-of-service requirements. We study the role of OPRs in two different network approaches: an OPR adopting the legacy of a telecommunication solution and a solution pertinent to the role and the functionality of an IP world. To benchmark the two scenarios, a multilayer, multigranular OPR architecture is presented and its potential to cost-effectively scale toward petabit-per-second throughput is justified by means of physical layer performance and power consumption estimations for each case. Furthermore, the blocking and frame-loss performance of the proposed OPRs is presented, proving the viability of the proposed solutions.

Next generation optical nodes: The vision of the European research project IDEALIST

As traffic demands become more uncertain and newer services continuously arise, novel network elements are needed to provide more flexibility, scalability, resilience and adaptability to todays optical networks. Considering these requirements, within the European project IDEALIST the investigation of elastic optical networks is undertaken with special focus on next generation optical node architectures. As an evolution of existent ROADMs and OXCs, these optical nodes will establish a new paradigm in which the network requirements will be efficiently addressed considering various emerging dimensions. In this article, we describe the drivers, architectures, and technologies that will enable these novel optical nodes. In addition, multivendor traffic interoperability, optical defragmentation, and node cascadability are also described as considerations in the node design.

Efficient burst aggregation for QoS‐aware slotted OBS systems

A crucial issue in slotted OBS systems is the slot-generating traffic aggregation strategy as it impacts heavily on efficiency. The longer the assembly time, the better the fill level, at the penalty of delay, which may affect sensitive services. To fulfil the QoS requirements of demanding traffic and achieve high efficiency whenever possible, two traffic priorities are employed and two different bounds on the aggregation delay are enforced. A slot aggregation strategy that creates a composite slot, improving utilisation gain while respecting service guarantees, is presented and evaluated in this paper. Under this approach a slot is launched whenever high-priority traffic cannot wait any longer but any low-priority traffic is accommodated as well, producing an efficiency gain in the sense that more traffic can be sent using the same number of slots. Copyright

Design and performance evaluation of a Programmable Packet Processing Engine (PPE) suitable for high-speed network processors units

In this paper, we present a Programmable Packet Processing Engine suitable for deep header processing in high-speed networking systems. The engine, which has been - fabricated as part of a complete network processor, consists of a typical RISC-CPU, whose register file has been modified in order to support efficient context switching, and two simple special-purpose processing units. The engine can be used in a number of network processing units (NPUs), as an alternative to the typical design practice of employing a large number of simple general purpose processors, or in any other embedded system designed to process mainly network protocols. To assess the performance of the engine, we have profiled typical networking applications and a series of experiments were carried out. Further, we have compared the performance of our processing engine to that of two widely used NPUs and show that our proposed packet-processing engine can run specific applications up to three times faster. Moreover, the engine is simpler to be fabricated, less complex in terms of hardware complexity, while it can still be very easily programmed.

HYDRA: A Scalable Ultra Long Reach/High Capacity Access Network Architecture Featuring Lower Cost and Power Consumption

This paper proposes HYbriD long-Reach fiber Access network (HYDRA), a novel network architecture that overcomes many limitations of the current WDM/TDM PON approaches leading to significantly improved cost and power consumption figures. The key concept is the introduction of an active remote node that interfaces to end-users by means of the lowest cost/power consumption technology (short-range xPON, wireless, etc.) while on the core network side it employs adaptive ultra-long reach links to bypass the metropolitan area network. The scheme leads to a higher degree of node consolidation and access-core integration. We demonstrate that HYDRA can achieve very high performance based on mature component technologies ensuring very low cost end-user terminals, reduced complexity, and high scalability.

Next generation optical network architecture featuring distributed aggregation, network processing and information routing

The ontology of communications is rapidly changing, shifting interest to machine-to-machine (M2M) interactions and the internet of Things (IoT). These are becoming vital for sustainability of social life and the revitalization of the economy providing the infrastructure to new production forms like distributed manufacturing, cloud robotics while becoming important to grid-based energy systems. Adding to them the voracious needs for data of the traditional broadband users, residential or business, together with the back/front hauling requirements of mobile operators, one is expecting a significant strain in the access. A multitude of heterogeneous access networks are emerging and the integration of them in a single platform ensuring seamless data-exchange with Data-Centres is of major importance. In this paper we describe HYDRA (HYbriD long-Reach fiber Access network), a novel network architecture that overcomes the limitations of both long-reach PONs as well as mobile backhauling schemes, leading to significantly improved cost and power consumption figures. The key concept is the introduction of an Active Remote Node (ARN) that interfaces to end-users by means of the lowest cost/power consumption technology (short-range xPON, wireless, etc.) whilst on the core network side it employs adaptive ultra-long reach links to bypass the Metropolitan Area Network. The scheme leads to a higher degree of node consolidation, network convergence and Access-Core integration. The proposed architecture can enhance performance while supporting network virtualization and efficient resource orchestration based on Software Defined Networking (SDN) principles and open access networking models.