随着若干新型云无线接入网（Cloud Radio Access Network, C-RAN）技术的出现，当今的网络环境发生了巨大变化，正经历快速转型。其中最重要的一类是基于以太网的C-RAN，为支持该技术，光以太网交换机等众多产品近期已推向市场。本文针对此类交换机的丢包率（Packet Loss Ratio, PLR）、时延和分组时延变化（Packet Delay Variation, PDV）开展性能分析。我们采用基于Simula离散事件建模（Discrete Event Modelling on Simula, DEMOS）的Simula工具——这是一种离散事件仿真的上下文类——来仿真直通式光以太网交换机在两类流量下的表现：高优先级（High priority, HP）流量和低优先级（Low priority, LP）流量。通过这种方式，本文对光以太网交换机的性能进行了定量评估。仿真结果表明，高服务质量体现在HP流量上，而LP流量的服务质量较低。因此，HP流量可用于传输以太网承载无线（Radio over Ethernet, RoE）流量，而LP流量可用于传输对时间不敏感的应用。研究还发现，在光以太网交换机中，HP流量的PDV等于最大尺寸LP流量的持续时间。
英文摘要：
With the advent of several new Cloud Radio Access Network (C-RAN) technologies, todays networking environment is dramatically altered and is experiencing a rapid transformation. One of the most important is Ethernet based C-RAN, in support of which many products such as optical Ethernet switches have recently appeared on the market. This paper presents the performance analysis of such switches with respect to Packet Loss Ratio (PLR), Latency and Packet Delay Variation (PDV). We employed the Simula based on Discrete Event Modelling on Simula (DEMOS), a context class for discrete event simulation to simulate a cut-through optical Ethernet switch under two types of traffics: High priority (HP) traffic and Low priority (LP) traffic. In this way, the paper evaluates the optical Ethernet switch performance quantitatively. The results obtained from the simulator showed that the high quality of service was reflected on HP traffic and the low quality of service in LP traffic. Hence, HP traffic can be used for transporting Radio over Ethernet (RoE) traffic while LP traffic can used for transporting time insensitive application. It is also found that HP traffic experiences a PDV equals to the duration of maximum sized LP traffic in Optical Ethernet switch.
The objective of this paper is to propose models enabling to study the behaviour of Ethernet switch for Networked Control Systems. Two scheduler policies are analyzed: the static priority and the WRR (Weighted Round Robin). The modelling work is based on Coloured Petri Nets. A temporal validation step based on the simulation of these modelling, shows that the obtained results are near to the expected behaviour of these scheduler policies.
Virtualisation first and cloud computing later has led to a consolidation of workload in data centres that also comprises latency-sensitive application domains such as High Performance Computing and telecommunication. These types of applications require strict latency guarantees to maintain their Quality of Service. In virtualised environments with their churn, this demands for adaptability and flexibility to satisfy. At the same time, the mere scale of the infrastructures favours commodity (Ethernet) over specialised (Infiniband) hardware. For that purpose, this paper introduces a novel traff…
Modern Ethernet switches support many advanced features beyond route learning and packet forwarding such as VLAN tagging, IGMP snooping, rate limiting, and status monitoring, which can be controlled through a programmatic interface. Traditionally, these features are mostly used to statically configure a network. This paper proposes to apply them as dynamic control mechanisms to maximize physical network link resources, to minimize failure recovery time, to enforce QoS requirements, and to support link-layer multicast without broadcasting. With these advanced programmable control mechanisms, st…
A 100 Mbps Ethernet link between a college campus and the outside world was monitored with a dedicated PC and the measured data analysed for its statistical properties. Similar measurements were taken at an internal node of the network. The networks in both cases are a full-duplex switched Ethernet. Inter-event interval histograms and power spectra of the throughput aggregated for 10ms bins were used to analyse the measured traffic. For most investigated cases both methods reveal that the traffic behaves according to a power law. The results will be used in later studies to parameterise models…
The need for optical parallelization is driven by the imminent optical capacity crunch, where the spectral efficiency required in the coming decades will be beyond the Shannon limit. To this end, the emerging high-speed Ethernet services at 100 Gbps, have already standardized options to utilize parallel optics to parallelize interfaces referred to as Multi-lane Distribution (MLD). OFDM-based optical network is a promising transmission option towards the goal of Ethernet parallelization. It can allocate optical resource tailored for a variety of bandwidth requirements and that in a fundamentall…
Ethernet-based Passive optical network (E-PON) is the key for next generation access networks. It must have the property of high efficiency, low cost, and support quality of service (QoS). We present a novel media access control (MAC) protocol that maximizes network efficiency by using dynamic bandwidth allocation (DBA) algorithm suitable for E-PON. This protocol minimizes packet delay and delay variation for high priority traffic to ensure QoS. Simulation results show excellent network throughput. Simulation results also show low packet delay and packet delay variation for high priority traff…
In this paper we discuss the implementation of an ONOS application that leverages Energy-Efficient Ethernet links between a pair of switches and shares incoming traffic among the link in the way that minimizes overall energy usage. As the straightforward solution can result in excessive traffic delay, we provide two alternative solutions to meet the demands of real time traffic arriving to the link aggregate. Experimental results show that our final application can keep low energy usage while meeting the demands of time-sensitive traffic, as long as the latter does not represent an excessive s…
We introduce the real-time multi-technology transport layer monitoring to facilitate the coordinated virtualisation of optical and Ethernet networks supported by optical virtualise-able transceivers (V-BVT). A monitoring and network resource configuration scheme is proposed to include the hardware monitoring in both Ethernet and Optical layers. The scheme depicts the data and control interactions among multiple network layers under the software defined network (SDN) background, as well as the application that analyses the monitored data obtained from the database. We also present a re-configur…
During the last few years, there has been plenty of research for reducing energy consumption in telecommunication infrastructure. However, many of the proposals remain unim-plemented due to the lack of flexibility in legacy networks. In this paper we demonstrate how the software defined networking (SDN) capabilities of current networking equipment can be used to implement some of these energy saving algorithms. In particular, we developed an ONOS application to realize an energy-aware traffic scheduler to a bundle link made up of Energy Efficient Ethernet (EEE) links between two SDN switches. …
As the demand for Internet expands significantly in numbers of users, servers, IP addresses, switches and routers, the IP based network architecture must evolve and change. The design of domain specific processors that require high performance, low power and high degree of programmability is the bottleneck in many processor based applications. This paper describes the design of ethernet packet processor for system-on-chip (SoC) which performs all core packet processing functions, including segmentation and reassembly, packetization classification, route and queue management which will speedup …
As new networking paradigms emerge for different networking applications, e.g., cyber-physical systems, and different services are handled under a converged data link technology, e.g., Ethernet, certain applications with mission critical traffic cannot coexist on the same physical networking infrastructure using traditional Ethernet packet-switched networking protocols. The IEEE 802.1Q Time Sensitive Networking (TSN) task group is developing protocol standards to provide deterministic properties on Ethernet based packet-switched networks. In particular, the IEEE 802.1Qcc, centralized managemen…
Federated Learning (FL) is a distributed machine learning (ML) type of processing that preserves the privacy of user data, sharing only the parameters of ML models with a common server. The processing of FL requires specific latency and bandwidth demands that need to be fulfilled by the operation of the communication network. This paper introduces a Dynamic Wavelength and Bandwidth Allocation algorithm for Quality of Service (QoS) provisioning for FL traffic over 50 Gb/s Ethernet Passive Optical Networks. The proposed algorithm prioritizes FL traffic and reduces the delay of FL and delay-criti…
Ethernet has become the next standard for automotive and industrial automation networks. Standard extensions such as IEEE 802.1Q Time-Sensitive Networking (TSN) have been proven to meet the real-time and robustness requirements of these environments. Augmenting the TSN switching by Software-Defined Networking functions promises additional benefits: A programming option for TSN devices can add much value to the resilience, security, and adaptivity of the environment. Network simulation allows to model highly complex networks before assembly and is an essential process for the design and validat…
The use of Ethernet switched networks usually involves best effort service. A recent effort by the IEEE 802.1/3 TSN group has sought to standardize the Ethernet data-link protocol such that it operates on a deterministic service in addition to the best effort service targeting Operational Technology applications, e.g., industrial control systems. This paper investigates the Cyclic Queueing and Forwarding (CQF) and Paternoster scheduling protocols in a typical industrial control loop with varying propagation delays emulating large scale networks. Our main findings for CQF and Paternoster are th…
The main issue in the optical transmission is switching speed. The optical packet switching faces many significant challenges in processing and buffering. The generalized multilevel protocol switching seeks to eliminate the asynchronous transfer mode and synchronous optical network layer, hence the implementation of IP over WDM (wave length division multiplexing). Optical burst switching attempts to minimize the need for processing and buffering by aggregating flow of data packets in to burst. In this paper there is an extensive overview on current technologies and techniques concerning optica…
High-speed photonic switching networks can switch optical signals at the rate of several terabits per second. However, they suffer from an intrinsic crosstalk problem when two optical signals cross at the same switch element. To avoid crosstalk, active connections must be node-disjoint in the switching network. In this paper, we propose a sequence of decomposition and merge operations, called conjugate transformation, performed on each switch element to tackle this problem. The network resulting from this transformation is called conjugate network. By using the numbering-schemes of networks, w…
This paper proposes a two-stage optical packet switch with second stage of recirculate switch of FDL to reduce the number of the FDL used in the switch for contention resolution. The contention resolution scheme with priority in packet releasing from FDL is tested in the two-stage switch for performance evaluation. Simulation result shows that zero packet loss rate achievable with ı< 0.8 for 32x 32 two-stage switch.
Relying on the flexible-access interconnects to the scalable storage and compute resources, data centers deliver critical communications connectivity among numerous servers to support the housed applications and services. To provide the high-speeds and long-distance communications, the data centers have turned to fiber interconnections. With the stringently increased traffic volume, the data centers are then expected to further deploy the optical switches into the systems infrastructure to implement the full optical switching. This paper first summarizes the topologies and traffic characterist…
