By Topic

Photonic Networks, 2009 ITG Symposium on

Date 4-5 May 2009

Filter Results

Displaying Results 1 - 25 of 25
  • Performance of Network Coding in Transport Networks with Traffic Protection

    Publication Year: 2009 , Page(s): 1 - 7
    Cited by:  Papers (3)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1485 KB)  

    Increasing traffic demands in the transport networks as well as declining revenues per bits require an increased transport efficiency. The concept of network coding has shown to increase the network throughput in capacity limited networks compared to traditional routing solutions. We show in this paper that applying network coding to certain reference transport networks with traffic protection reduces the required transport resources thus improves the network efficiency. Besides investigating the achievable resource savings we elaborate constrains and limiting factors for the performance and application of network coding in transport networks with a 1+1 path protection. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dispersion Monitoring for Advanced Modulation Formats using Nonlinear Detection

    Publication Year: 2009 , Page(s): 1 - 7
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (225 KB)  

    In high bit rate transmission systems beyond 10 Gbit/s (such as 100 Gbit/s Ethernet) static dispersion compensation is not sufficient to guarantee reliable links. Dispersion monitoring is necessary for adaptive dispersion compensators. Dispersion estimation by means of nonlinear detection is a promising approach that was studied so far for amplitude modulation formats like NRZ and RZ. In this paper, we show that this technique can also be used for modulation formats like DPSK and DQPSK. The behavior of these modulation schemes in the presence of chromatic dispersion is shown and the output of the nonlinear detector is explained using numerical simulations and analytical calculations. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Scalable Architectures for 100 GbE Packet Processing

    Publication Year: 2009 , Page(s): 1 - 7
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (100 KB)  

    This paper addresses the question, how device technologies and architectures scale towards 100G packet processing. The evolution of field programmable gate array (FPGA) and application-specific integrated circuit (ASIC) technology is analyzed in terms of throughput, interfaces, density, and power. Moreover, the 100 Gigabit Ethernet (100 GbE) component ecosystem is subject of a detailed review regarding functional split and interfaces. Domains are identified that cannot keep up with the increased throughput demand, and therefore limit higher integration. It is concluded that device technology has to improve significantly in the strive for economical and energy-efficient 100 GbE networking. In the meantime, techniques such as caching, pipelining, and load balancing open viable paths towards 100 GbE packet processing. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Packet Processing at 100 Gbps and Beyond - Challenges and Perspectives

    Publication Year: 2009 , Page(s): 1 - 10
    Cited by:  Papers (3)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (664 KB)  

    The continuous growth of traffic volumes steadily raises the throughput requirements on the network infrastructure. Additionally, a transformation of the classical TDM-based backbone networks to packet networks with Carrier Ethernet as the target technology occurs. The standardization process of 100 Gbps Ethernet is under way. This not only poses big challenges to transmission but also to packet processing technologies. However, recent announcements from network processing unit (NPU) vendors promise that packet processing at 100 Gbps is feasible. The big question for system manufacturers now is, whether this trend will continue and finally lead to 1 Tbps packet switching, or whether there are technological roadblocks that inhibit this development path. In this paper, we address this question and identify packet processing performance, packet buffer throughput, chip-to-chip interface speed, and power dissipation as the most critical factors. We discuss their limiting factors as well as architectural and technological trends that can further increase their performance. Based on these investigations and extrapolating anticipated technological advances we expect that 1 Tbps packet processing and switching could be introduced in the network within several years. Since this, however, not only depends on technological but also on economical factors, we show how slight modifications of the network architecture and protocols could alleviate some implementation complexities and thus reduce the overall cost. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 100GET¿OCTET ¿ Higher Speed Networking

    Publication Year: 2009 , Page(s): 1 - 3
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (940 KB)  

    Future telecommunications networks have to be able to transport highly increased packet traffic volumes at strongly reduced cost per bit basis. Due to exponentially growing traffic volumes and almost flat rate pricing, the needed capacity has to be provided at least at the same level of total costs, or even below, compared with today. At the same time, it is necessary for converged networks to maintain the performance and resiliency at carrier grade level. The simplification of the network architecture and the introduction of higher speed interfaces point towards that direction. A cost efficient transport has to be founded and performed in those layers with respect to the optical and packet network, which offer lowest cost per bit performance at carrier grade quality. Furthermore, a traffic aggregation towards terabit transmission for scalable networks is expected, increasing both the speed up to 100 Gbit/s standardized interfaces and beyond and the utilization of the optical transmission path of fibers by increasing the spectral efficiency. In this paper the status of standardization of higher speed interfaces is reviewed, first 100 Gbit/s transmission experiments over Deutsche Telekom field installed fibers within the BMBF 100GET OCTET project are presented, and the layered architecture and functionality of future photonic and packet networks are discussed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Backup Path Reprovisioning and Activation Planning with Differentiated Dual-Failure Restorability in WDM Mesh Networks

    Publication Year: 2009 , Page(s): 1 - 5
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (233 KB)  

    A general integer linear programming (ILP) model, which covers all possible second-failure flow recovery cases, is proposed in this paper to capture the structure of the global backup path reprovisioning scheme triggered after the first-failure flow recovery. Backup capacity contention in face of a new failure is resolved by incorporating a priority activation planning process into the model to achieve differentiated dual-failure restorability for two classes of flows. Numerical results show that without increasing the overall network-wide dual-failure non-restorability, the average network-wide dual-failure non-restorability of high class flows is relatively lower than that of low class flows by over 69% with the proposed differentiated backup path reprovisioning and activation planning (RAP) scheme. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Frame Assembly in Packet Core Networks ¿ Overview and Experimental Results

    Publication Year: 2009 , Page(s): 1 - 8
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1477 KB)  

    Literature has proposed Frame Assembly and its variants multiple times to cope with the ever increasing switching density in consequence of increasing link rates. Nevertheless, state-of-the-art networks do not implement and apply it. Skepticism of practitioners and investors regard not only the effective gain of frame switching, but also questions of control, interfacing and performance impact on the existing Ethernet/IP infrastructure. We present an operational prototype network with frame assembly in its core that seamlessly interfaces to existing Ethernet technology and seamlessly integrates to a standard conform GMPLS control plane. We show the manageable additional effort of assembly at the network edge, the direction how to integrate such network into existing control structures, but also the limited and well controlled impact of assembly on the timing of client application. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • NG-PON ¿ TDMA or WDM?

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (932 KB)  

    Today, there is a major trend in broadband metro access networks towards FTTx (Fiber-To-The-x, where x stands for Home (H), Building (B), Curb, or Cabinet). FTTx can be based on active Point-to-Point/Ethernet, or Passive Optical Networks (PON). Many agree that PON-based FTTH/FTTB will win mid- to long-term over other deployment scenarios because of high bandwidths at comparatively low cost. However, there is also debate as to whether or not the future passive access network should be based on Time-Domain Multiple Access (TDMA, as known from GPON and EPON) or on WDM/WDMA (Wavelength-Domain Multiple Access). In this paper, we analyze requirements for the unified next-generation access and backhaul infrastructure. We conclude that this infrastructure should be based on WDM rather than TDMA in order to accommodate future band-width requirements whilst at the same time supporting backhaul and business access applications most efficiently. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Low-energy demarcation device for monitoring FTTx networks

    Publication Year: 2009 , Page(s): 1 - 3
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (672 KB)  

    In a passive optical network, network availability as well as network security will become increasingly important. Traffic-transparent energy-autarkic supervision units located at demarcation points between network segments (so-called demarcation devices, DD) would fill this need. The DD respond to a low-duty cycle low-bitrate control signal which the operator¿s central office (CO) superimposes to the optical data stream sent to the subscribers. Most of the time the DD stay in a low-energy sleep mode, but awake during very short time intervals for communicating with the CO. In sleep mode, synchronism with the CO is lost. Our new low-energy medium-access control (LE-MAC) protocol assures a fixed ¿rendezvous¿ time where all DD can individually respond to the CO. In a small network we show the feasibility of the concept and the functionality of the LE-MAC protocol. The mean current consumption of a DD is less than 1 muA corresponding to 3 muW. With a single inexpensive lithium ion battery, such a DD could operate for more than 10 years. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 100Gb/s DPSK-3ASK Modulation Format for Metro Networks: Experimental Results

    Publication Year: 2009 , Page(s): 1 - 3
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (189 KB)  

    We assembled an experimental setup to investigate the optical signal characteristics of a multi-level 100Gb/s modulation format that demonstrates the possibility to be a low-cost option for higher speed interfaces in metro networks. The modulation format which we are naming DPSK-3ASK is comprised of phase-shift keying as well as amplitude-shift keying to produce 2.5bits/symbol. This encoding allows the system to be realizable with 40Gb/s components. To generate the three amplitude levels for the ASK modulation, we use a 40Gb/s bit-error ratio tester and a multiple-path RF line to drive a commonly available 40Gb/s modulator which is biased at the quadrature of its transfer function. We place a second 40Gb/s modulator biased at the null of its transfer function for phase modulation. We detect the six symbols in this modulation format by first using an optical splitter to direct the signal to two separated paths: an ASK path comprising of a single photo-detector for direct detection, and a PSK path that utilizes a delay-line interferometer followed by a balanced detectors. We measure the bit error rate for the two ASK eyes and for the PSK eye, as well as show the noise tolerance of the modulation format by coupling amplified spontaneous emission noise together with the signal. We discuss practical challenges with the setup, and list possible solutions. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 8x107 Gbit/s NRZ-VSB DWDM Field Transmission over 500 km SSMF

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (941 KB)  

    We present the results of an 8x107 Gbit/s ASK-NRZ-VSB DWDM field transmission trial in the fiber network of Deutsche Telekom. The 8x107 Gbit/s DWDM signal was generated by two Mach-Zehnder modulators each modulating a group of four wavelength on a 200 GHz grid for decorrelation of the neighbouring DWDM channels. Both wavelength channel groups are combined by a 100 GHz interleaver, which additionally converts the NRZ signals by vestigial-side-band (VSB) filtering to NRZ-VSB signals with 1 bit/s/Hz spectral efficiency. The transmission link consisted of 6 spans of standard single-mode fiber with an overall link length of 500 km. Pure EDF-amplification was used to compensate for the fiber loss. The link included two spans with high attenuation (~29 dB) and two spans with high PMD resulting in an overall link PMD of ~8 ps. Within the wavelength band covered by the DWDM channels the DGD was in the range of 3 to 15 ps, which corresponds up to the 1.6-fold of the bitperiode of the 107 Gbit/s OOK signal. Therefore application of a optical PMD compensation was mandatory for the transmission trial. After DWDM demultiplexing and direct detection of the 107 Gbit/s data signal in the ETDM receiver the BER of all tributaries of all DWDM channels was measured. We measured an average BER of 7x10(exp -4) and achieved stable operation . View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • 1 Tb/s Transmission on a Single Wavelength Channel for Future Terabit Ethernet

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1498 KB)  

    We have performed an ultra-high speed transmission experiment at a data rate of 1.07 Tb/s on a single wavelength channel. In contrast to previously reported single channel transmission experiments at similar data rates, our transmission system is based exclusively on semiconductor pulse sources and does not require sophisticated schemes for RZ-pulse compression. Besides a 13.375 GHz semiconductor mode-locked laser, the 1.07 Tb/s transmitter comprises a simple scheme for moderate pulse compression in an Erbium doped fiber amplifier. We employ alternating polarization RZ-DQPSK modulation and optical time division multiplexing to generate the data signal for transmission. In the 1.07 Tb/s receiver we use a fiber based high-speed optical time-division demultiplexer which is driven by a 53.5 GHz monolithically integrated mode-locked laser diode. Further demultiplexing at lower bit rates is performed in the electrical domain after differential optical demodulation of the data signal. Taking into account a 7% overhead for forward-error correction, we achieve error-free transmission of 1 Tb/s data over a transmission length of 480-km dispersion managed fiber. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Spectrum-based Frequency Offset Estimation for Coherent Receivers

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1576 KB)  

    We propose applications of well-known algorithms for non-data aided spectrum-based frequency offset estimation in optical coherent receivers. The algorithms are independent of the modulation format and robust against channel distortion. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Analysis of an ASIC-based Coherent Polarization-Multiplexed QPSK Receiver and Different Receiver Frontends

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1318 KB)  

    In this paper we present the characterization results of an optical coherent quadrature phase shift keying (QPSK) receiver with an application-specific integrated circuit (ASIC) based backend unit. Additionally we investigate the performance of an integrated optical coherent receiver frontend (ICR) and compare it against a fiber-pigtailed 90deg optical hybrid. After characterization of the CMOS ASIC, we measure and analyze single polarization QPSK transmission at 2.5 Gbaud with different optical receiver frontend setups. Also we measure the polarization-multiplexed QPSK transmission and discuss the possible reasons for reduced sensitivity compared to the single-polarization transmission. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Advantage of Reach-Dependent Capacity in Optical Networks Enabled by OFDM

    Publication Year: 2009 , Page(s): 1 - 5
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1277 KB)  

    Orthogonal frequency-division multiplexing allows implementing reach-dependent transmission capacity. The advantage of such an approach in optical networks is presented for two different network scenarios. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Analytical Approach to Cost Optimal Network Design

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (801 KB)  

    The growth of the Internet and its applications is built on the success of optical transport technologies, enabling the sustained bandwidth growth observed. The advent of fibre transport in the access and aggregation networks imposes new challenges on network operators. Most networks were optimised according to the limitations of copper twisted-pair or coaxial cables, demanding stringent limits for the distances between the end user, switches and local exchanges. The high bandwidth-demand product of optical fibres allows operators to give up their old infrastructure by benefitting from unamplified transmission distances of 40+ kilometres. Multiple technologies exist to satisfy the needs, either based on a purely optical transport solution like fiber-to-the-home (FTTH), or hybrid approaches leveraging for the last mile on the existing copper infrastructure in solutions like fiber-to-the-curb (FTTB). Larger transmission distances, however, translate in a higher number of end users whose traffic needs to be terminated in the network switches and IP gateways. In this paper we present an analytical approach on identifying cost-optimal solutions for such new network architectures, stretching from the access to the core network. The large number of technological and architectural options for a network redesign renders a simulation-based approach impossible. The analytical model used here provides immediate results on the capital expenditure required for a certain architecture, taking into account transport and switching costs in the access, aggregation and core of the network. The analytical model trades accuracy in detail for tractability and understanding of the relevant effects governing overall network design. Prinipical and numerical results obtained by applying the model to a fictive Germany-wide network are presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Opex Considerations for Optical Switch Upgrades: A Multiperiod Point of View

    Publication Year: 2009 , Page(s): 1 - 7
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (241 KB)  

    In order to provide the necessary capacity on the optical layer, one possibility is to light more fibers. Thereby the fiber nodal degrees of the optical switches are increased. In this paper we propose traffic affecting and non-affecting upgrading types for optical switch models, focusing on operational expenditures (Opex). Opex costs for node upgrades provide a small cost proportion of the overall costs. However, our results indicate that the correlated penalty costs for traffic interruption due to traffic affecting upgrades have considerable impact on the planning result. Two long-term planning approaches provide an upper and a lower bound of the planning optimality and illustrate the necessity of including forecast knowledge to the optimization. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Generic Algorithm for CAPEX-Aware Multi-Layer Network Design

    Publication Year: 2009 , Page(s): 1 - 8
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1591 KB)  

    Capital expenditure (CAPEX), i.e. equipment cost, is a decisive criterion for the design of a new network infrastructure. The design of efficient transport networks requires the interconnection of multiple technologies which form separate layers with their own networking view and routing. This leads to the notion of multi-layer networks. CAPEX-aware design of such networks requires algorithms which propose multi-layer hardware configurations that are able to carry a given set of traffic demands at minimal CAPEX.We present a generic algorithmfor CAPEX-aware multi-layer network design as well as a computationally viable implementation and perform evaluations on realistic network topologies. The underlying CAPEX and multi-layer technology models are explained in detail. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Electronic Equalization of FBG Phase Ripple Distortions in 43 Gb/s WDM Systems

    Publication Year: 2009 , Page(s): 1 - 7
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (2217 KB)  

    Within this paper we investigate in a first step the transmission performance for a 43-Gb/s wavelength division multiplex (WDM) system with channelized and broadband fiber Bragg gratings (FBGs) for dispersion compensation by statistical simulations. The results demonstrate the feasibility and limitations of replacing dispersion compensating fiber (DCF) in 43 Gb/s systems by FBGs and of upgrading a 10.7 Gb/s system with FBG dispersion compensation to 43 Gb/s. Phase ripple distortions and the narrow band filtering of the channelized FBG extend the OSNR requirements or lead to reduced possible transmission lengths. To overcome these limitations, linear feed-forward-equalizers (FFE), combinations of feed-forward and decision-feedback (FFE-DFE) equalizers, and nonlinear FFE-DFEs are introduced. We show that the OSNR penalty due to filtering and ripples can be significantly reduced by using nonlinear equalizers. Investigated modulation formats are NRZ-ASK (amplitude shift keying with non-return-to-zero pulse shape) and optical duobinary (ODB). View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Demonstration of an InP-based Electrical 1:2 Demultiplexer in a 107 Gb/s OOK System

    Publication Year: 2009 , Page(s): 1 - 5
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1671 KB)  

    The development of transmission technology for 100 Gigabit/s Ethernet systems is an active field of research worldwide. The development is concentrating mainly on low-baud-rate systems (up to 50 Gbaud) in conjunction with higher-order modulation formats (e.g. QPSK). These systems have the benefit of high spectral efficiency and robustness against chromatic dispersion or PMD. On the other hand they require complex, less cost-efficient transmitter and receiver architectures. A possible alternative, in particular for short reach, are higher-baud-rate systems (100 Gbaud). These systems have a simpler transmitter and receiver architecture which could result in cost-efficient realizations. However, in order to realize these architectures, compact and integrated solutions for the key components (optical-toelectrical-, electrical-to-optical-conversion, electrical amplification and processing) are required. Within the European project HECTO, funded under FP6, such key components are developed, based on InP technology. The use of InP technology offers the potential advantage of higher speed, higher breakdown voltages and monolithic integration (e.g. demultiplexer, photodiode and amplifier) on a single chip. In this paper, we report on the application of an InP-based electrical 1:2 demultiplexer (DEMUX), developed within HECTO, in a 107 Gb/s (100 Gb/s + FEC overhead) OOK system. The receiver was based on pure electrical timedivision multiplexing. It comprised a high-bandwidth photodetector, also developed within HECTO, and the DEMUX. The transmitter used optical time division multiplexing (OTDM) techniques. Measurements of eye diagrams, bit-error rates for 53.5 and 107 Gbit/s back-to-back, as well as investigations on the tolerance of the DEMUX to decision time and threshold (V-curves) are presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Experimental Analysis of 10Gb/s Optical OFDM Based on Intensity Modulation with Direct Detection

    Publication Year: 2009 , Page(s): 1 - 5
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1486 KB)  

    It is the purpose of this contribution to experimentally verify recent theoretical outcomes related to the trade-off between spectral efficiency and receiver sensitivity on one hand and the interplay of modulator biasing and modulator driving voltage swing for optimum receiver sensitivity on the other hand. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Heuristic approach to forecast the number of wavelength services in future OTN networks

    Publication Year: 2009 , Page(s): 1 - 7
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1403 KB)  

    The continuous evolution of photonic technologies and the first experience with photonic switches deployed in the field motivate operators to consider photonic networking in their next generation networks. For the strategic planning of multi-layer networks a realistic traffic estimate of customer services is essential in order to achieve an optimized network design meeting the operational and economical requirements. Some types of customer services like e.g. ï¿¿ï¿¿ï¿¿digital leased linesï¿¿ï¿¿ï¿¿ and ï¿¿ï¿¿ï¿¿broadbandï¿¿ï¿¿ï¿¿ have a history of several years, sufficient experience and analystsï¿¿ï¿¿ï¿¿ data is available to allow a plausible forecast. The situation is different with customer lambda services. There has been a relatively small market of customer and carrier lambda leased lines served by DWDM transport. Since 2005 there are indicators that this market started growing strongly and that lambda services will be a significant part of the next generation OTN traffic, but the currently available data hardly enables a reliable estimate in particular for 40G and 100G services. This paper proposes a method to forecast the number of lambda services considering the growth of existing services as well as the emergence of future higher capacity services based on very few input parameters: the number of services and the share of the different capacity classes in the reference year, the expected annual total traffic growth, and optionally the expected annual total growth of the number of services. The calculation model is based on the distribution function of node and link capacities empirically detected during several transport network studies, which coarsely follows a negative exponential curve. Growing the traffic according to this distribution and mapping it to the lambda capacity classesï¿¿ï¿¿ï¿¿ grid provides the evolution of the different services over time. Some forecast examples are shown regarding the German optical leased lines market considering different start and growth assumptions. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Increasing Packet Sizes to Mitigate Performance Issues in High-Speed Packet Networks

    Publication Year: 2009 , Page(s): 1 - 8
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1258 KB)  

    Emerging packet-switched transport networks face a continuous growth of link bitrates. Without counter-measures, this translates into a proportional increase of the packet rate and may turn packet processing in network nodes into a bottleneck. In this paper, we investigate one approach to reduce the processing load for a given line rate: increasing the maximum size of packets exchanged between end systems. We first present the fundamental mechanisms along with side effects of increased packet sizes. Then we discuss the dependence of the achievable packet rate reduction on application and protocol stack. Based on a traffic model from the year 2000 and current measurements, we finally evaluate the reduction for web traffic and the Internet protocol stack. Results show that today¿s web traffic enables higher packet rate reductions than traffic of 2000. A fourfold packet size increase now allows to cut the packet rate by two thirds. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Deduction of Video Quality Degradation in Transport Networks

    Publication Year: 2009 , Page(s): 1 - 7
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1216 KB)  

    It is already known that significance marking of IP video packets in conjunction with a corresponding evaluation in the transport network considerably protects the quality of videos against degradation from network congestion. The focus of this paper therefore is the investigation of possibilities to take benefit from such significance marks. In particular the interest is to estimate the video quality in the network without having to spend considerable effort for decoding or even decrypting of the video stream. While the merit of significance markings has already been proven by simulation, now the benefits have been demonstrated by experiments on actual transport network hardware. Thereby, network element typical Ethernet performance counters registering the amount of received and lost bytes have been modified for differentiation by the significance marks. Those counters, retrieved per second, were compared with the quality of the video. It is discussed how to construct a metric for video quality, and the role of a suitable definition of an objective reference for the calibration of this metric: The metric has to regard specifics of the video compression, video encoding and transport network characteristics. Results from these experiments are provided and challenges for further refinement and implementation are pointed out. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Next generation networking with 100G/100GbE

    Publication Year: 2009 , Page(s): 1 - 5
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (760 KB)  

    This paper provides a brief survey of the motivators behind IP traffic growth, and the locations in today¿s networks that are likely to be exposed to aggregated bandwidth, reaching and exceeding 100G. We will identify the major application spots for 100G technology, looking at TDM- and packet-based transport networks as well as data centers, with an analysis of the adherent functional differences. Given other economical and technical considerations, the demand for this challenging technology varies with available alternatives. Technical scalability limits make feature reductions of early, purpose-built implementations necessary. It is concluded that increasing bandwidth demand will force data centers to be first implementing 100G technology. Core- and metro (TDM-)transport networks face specific challenges in compatibility with deployed fiber base and available networking capabilities at 100G. Data networks that forward traffic at Layer 2 with technologies such as T-MPLS/ MPLS-TP can alleviate hurdles of poor-scaling packet processing tasks e.g. by avoiding IP address lookups, DPI and session-based QoS in IPv4/IPv6 routed networks, thus improving 100G product realization feasibility. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.