Transport Networks Introduction PDF

INTRODUCTION TO TRANSPORT NETWORK 45 TRANSPORT NETWORK: Reliable aggregation and transport of any client traffic type, in any scale, at lowest cost per bit. Requirements: - Scalability: ability to support any number of client traffic instances whatever...

INTRODUCTION TO TRANSPORT NETWORK 45 TRANSPORT NETWORK: Reliable aggregation and transport of any client traffic type, in any scale, at lowest cost per bit. Requirements: - Scalability: ability to support any number of client traffic instances whatever network size, from access to core (Layering, Partitioning) - Multi-service: ability to deliver any type of client traffic (transparency to service) - Quality: ability to ensure that client traffic is reliably delivered at monitored e2e performance (connection oriented, strong OAM, traffic engineering, resource reservation). - Cost-Efficiency: keeping processing complexity as low as possible and operations easy. (CAPEX, low protocol complexity and OPEX, unified management and control access) o CAPEX: capital expenditures creating future benefits (equipment, property, industrial buildings, networking software…) o OPEX: operating expenditures (operation costs, license fees, maintenance and repairs, insurance, advertising, supplies, utilities…) EVOLUTION OF NETWORKS - First digital transmission 1962 (Bell Labs): use of Pulse Code Modulation (PCM, patented in france in 1938) and Time Division Multiplex (TDM) techniques. o PCM goal was convert analogue voice telephone channel into a digital one, band-limited signals in the range of 300-3400Hz. Involves 3 phases: sampling (Nyquist), quantization and encoding. o BW = 4 KHz -> fs = 8 KHz. 8 bits to codify each sample. T = 1/125us. Bitrate obtained: 64Kbps o To reduce number of physical connections, communications carriers employ multiplexing. To enable multiple simultaneous voice conversations to be routed onto a common circuit -> TDM Technique o PCM/TDM allows several digital voice channels to be transmitted over the same media on a time divided basis. The multiplexer assembles a full character from each data source into a frame for transmission, byte-by-byte or channel-by-channel multiplexing. Slotted medium of 64Kbps/chnl. The resulting channel is the 1st level carrier: E-carrier (EUR) or T-carrier (USA) - Whichever transmission technology is used to transport voice channels, it must be able to transport one sample of each channel every 125us. The signals of all  lOMoAR cPSD| 9558445 hierarchical levels of PDH and SDH are organized in frames of the same duration (125us), independently of the number of transported channels. In this way, each byte is a specified position within the frame can carry one telephone channel or equivalently a digital channel of capacity 64kbps. PDH HIERARCHY To transport more than 30 channels, the ITU-T defined the hierarchy PDH: the higher hierarchical levels are obtained by multiplexing N lower levels frames within a frame whose nominal transmission rate is more than N times that of the lower level. The level 1 of the hierarchy is done Byte by Byte, but the higher levels are carried out bit-by-bit. Based on a plesiochronous (means almost synchronous) operation mode, with no synchronization network and clocks whose values fluctuates around a nominal frequency value (synchronism: setting up a common block between origin and reception nodes in order to correctly recover transmitted signals) - Higher levels multiplexers receives frames from lower level multiplexers with clocks whose value fluctuates around a nominal frequency value within certain margins of tolerances. The margins are set by the ITU-T recommendations for each hierarchichal level. The justification mechanism is needs for dealing with these fluctuations. Limitations of Plesiochronous Networks Plesiochronous ANSI and ETSI hierarchies were not compatible. There were no standards defined for rates over 45Mbps in T-carrier (USA) and over 140Mbps in Europe PDH. Their management, supervision and maintenance capabilities are limited, as there are no overhead bytes to support these functions. Different manufacturers of plesiochronous equipment could not always be interconnected, because they implemented additional management channels or proprietary bits rates, so in PDH it was not possible to create higher bits rates directly. Access to 64Kbps digital channels from higher PDH hierarchical signals requires full de- multiplexing: the use of bit-oriented procedures removes any trace of the channels and a lot of multiplexers and de-multiplexers are needed. These limitations means that it was necessary to design a new transmission architecture to increase flexibility, functionality, reliability and interoperability of networks->SDH. SDH INTRODUCTION The Synchronous Digital Hierarchy (SDH) is an ITU-T universal standard which defines a common INTRODUCTION TO TRANSPORT NETWORK 45 and flexible communication architecture to transport telecommunication services. - SONET (Synch. Optical Network) is today a subset of SDH, promoted by ANSI and developed at Bellcore Labs in 1985. In 1988, broadband-ISDN was created for transport simultaneously data, voice, video and multimedia over common transmission infrastructures. ATM was selected for the switching layer, and SDH for transport at the physical layer. SDH was first introduced into telecom. networks in 1992. It’s based on overlaying a synchronous multiplexed signal onto a light steam transmitted over fibre-optic cable and also for use on radio relay links, satellite links and electrical interfaces between equipment. SDH advantages: - Provides the definition of a flexible architecture capable of accommodating future applications with a variety of transmission rates (flexible transport of pleosynch and synch signals) - Enable transmission over multiple media, so they allow internetworking between different manufacturers by means of a set of generic standards and open interfaces. - Basic operations as multiplexing, mapping or alignment are byte oriented, to keep transported elements identifies through the whole transmission path. - All the nodes must be fed with the same master clock: synchronisation enables us to insert and extract tributaries directly at any point and at any bit rate, without delay or extra hardware. - Provides an overhead permitting operation, administration and maintenance (OAM) functions, which are essential to enable a centralized management. - Provides mechanisms to protect the network against link or node failures, to monitor network performance and to manage network events (SDH Resilience). - Scalability: transmission of rates of up to 40Gbps, making SDH a suitable technology for high-speed trunk networks. Some SDH Features SDH Network Elements - Regenerators: Se sincronizan usando la señal recibida y remplazan los bytes RSOH antes de retransmitir la señal. MSOH, POH y la carga útil no se altera. - Multiplexers para insentrar y extraer datos en muestras sincronizadas o LTM: combina plesióncronos y señales de entrada síncrona en una señal STM-N de mayor velocidad de bits. o LM: Señales síncronas multiplexadas/demultiplexadas STM-N dentro de STM-M (M>N). o ADM: Señales plesincronas y síncronas de bajo bit rate se pueden extraer o insertar en el bit SDH de alta velocidad de flujo. - Digital Cross-Connects (DXC): Permitir el mapeo de las señales tributarias de la PDH en contenedores virtuales, así como la conmutación de varios contenedores. El tráfico conmutado puede ser tanto corrientes SDH como afluentes seleccionados. Topología - Ring: se utilizan con frecuencia para construir arquitecturas de tolerancia a las fallas. La ventaja principal es su capacidad de supervivencia. Hierarchical Master-Slave Synchronization - A master clock synchronizes the slaves clocks, directly or indirectly according to a tree topology, and these are organized in two or more hierarchical levels.  lOMoAR cPSD| 9558445 - Protection mechanism against link and clock failures are planned through alternate synchronization routes, not only between parent and son clocks, but also between brothers, or even uncle and nephew. Elements of this architecture - Primary Reference Clock (PRC). Represents an autonomous clock or a clock that accepts reference synchronization from radio or satellite (GPS or Loran-C) Caesium atomic clock. - Slave clocks: SSU (Synchronization supply unit) o BITS (Building Integrated Timing Supply). There are transit and local nodes: SSU-T and SSU-L - Clocks at NE: SEC (SDH Equipment Clock) - Synchronous Ethernet has the same synchronization network model than SDH. - Higher bit rates over STM-1: Through the hierarchy. - Contiguous concatenation: Allows bit rates in excess of the capacity of the containers. The payload can be distributed to several containers. Creates big containers that cannot split into smaller pieces during transmission. o VC-4-4c (599.040 Mbps), VC-4-16c (2,396.160 Mbps), VC-4-64c (9,584.640 Mbps), VC-4-256c (38,338.560 Mbps). SDH Limitations - Transport different signal rates: other bit rates different to given by virtual containers (VC-x and contiguous concatenation) are not possible. INTRODUCTION TO TRANSPORT NETWORK 45 - The necessity to find one simple encapsulation method that was capable of accommodating any data packet protocols. - Bandwidth efficiency: the need to use bandwidth accurately Next generation SDH (NGN-SDH) - Virtual Concatenation (VCAT) to provide more flexibility in matching the bandwidth of the client signal. o Resolves the granularity problem of SDH adapting transmission speed to user requirements by using virtual concatenation. ▪ User data mapped to groups of virtual containers. Inverse multiplexing (G.805) o Optimizes the use of SDH network ▪ Virtual Concatenation offers the user a granular bandwidth choice, optimizing the use of network resources (better efficiency of the SDH network) o Transparency in the SDH network ▪ Individual VC are beared as traditional virtual containers. ▪ Core nodes are transparent to VCAT. ▪ End nodes must support VCAT functionalities. ▪ Receiver node reassembles the user frame and must compensate the delay differences of each path. Delay correction has a maximum limit of 512 ms. Suitable for continental networks (in fact, 100 ms is more than sufficient). - Link Capacity Adjustment Scheme (LCAS) to provide the ability to change the size of a VCAT signal. o G.7042. Provides soft protection and a mechanism for load sharing. Is an extension of virtual concatenation o Designed to manage the bandwidth allocation of a VCAT path. LCAS can add and remove members of a VCG that control a VCAT channel. Provides the ability to change the size of a VCAT signal. ▪ Dynamic bandwidth Allows bandwidth changes during the service BW can be managed adding or dropping VC of VCG Asymmetric Configurations. o Protection and failure tolerance ▪ Increases availability of VC from failures or changes ▪ Automatically decreases link capacity if a VC path has a failure, increasing when repaired. - Generic Framing Procedure (GFP) used for mapping packet-based signals into the constant bit-rate SDH signals.  lOMoAR cPSD| 9558445 G.709 OPTICAL TRANSPORT NETWORK - OTN technology enables multiple networks and services, such as legacy SONET/SDH, to be combined seamlessly into a common infrastructure for data, voice, video and storage applications -OTN or “digital wrapper” was created to combine the benefits of SDH/SONET technology with the bandwidth expansion capabilities offered by DWDM NETWORK EVOLUTION For decades, large high-speed backbone networks have been circuit-switched networks, based on TDM running SDH/SONET. Around 2012 - TDM transitioning to Ethernet. The primary traffic type is bursty - Peer-to-peer computing and cloud computing as the new models - Internet-based video applications putting demand on bandwidth - The explosion of BW demand -> entering the ZettaByte (ExaByte = 109 GB, ZettaByte = 1000 EB) - With the escalation in demand for video, data and other bandwidthhungry applications, the limitations of circuit-based systems have become readily apparent. o Underutilized or “wasted” bandwidth using TDM o TDM is expensive to operate o SDH capped at 40 Gbps. Evolution to packet switching driven by: - Growth in packet-based services (L2/L3 VPN, IPTV, VoIP, etc.) - Desire for bandwidth/QoS flexibility New packet transport networks need to retain same operational model An MPLS (Multi Protocol Label Switching) transport profile being defined at IETF (in collaboration with ITU-T) IP/MPLS has the flexibility and scalability to use that capacity most efficiently. MPLS-enabled core networks brings packet-awareness - MPLS is a method for providing connection oriented services for variable length frames without regard to their type: whether they be IP packets or native ATM, SDH or Ethernet frames. TECHNOLOGY FOR PACKETIZED NETWORKS - PSEUDOWIRE (PW): A mechanism that emulates the essential attributes of a native service while transporting over a packet switched network (PSN). Supports many L1&L2 services over a common packet switched network infrastructure. - PW+IETF -> MPLS/VPLS, emulates LAN over an MPLS network. - PW+IEEE -> Provider Backbone Transport (PBT). Tunnels interconnect Ethernet access clouds. - PW+ITU-T -> Transport MPLS (T-MPLS) - MPLS Transport-Profile (MPLS-TP): ITU-T and IETF convergence. INTRODUCTION TO TRANSPORT NETWORK 45 PACKET TRANSPORT CHARACTHERISTICS MPLS vs MPLS-TP EVOLUTION IN THE CLOUD ERA REQUIREMENTS IN THE CLOUD ERA The migration to cloud is leading to massive changes in how communications networks are built and operated. We will need: - Capacity scale - Networks and services agility (rapid reconfiguration and automation) - Openness (interoperability across domains, layers and vendors). The software-centric approach deploys faster, scales immediately, reduces capital expenditure, and simplifies network operations.  lOMoAR cPSD| 9558445 So, they appear new technologies: Software-Defined Networking (SDN) and Network Functions Virtualization (NFV). - Virtualization as enabler - Use of COTS (commercial off the shelf) reducing CAPEX - Flexible, automated and programable networks to reduce OPEX and delivering services faster to market (zero-touch management, Artificial Intelligence as a selflearning algorithms to adapt and execute in specific network operating situations dynamically). - Network slicing: Using SDN and NFV it will be possible to configure the type of network that is required by each type of service, over the same hardware with different software. SEGMENT ROUTING WITH SDN NEW NEED FOR WAN CONNECTIONS Enterprises built private data centers and then connected these assets with private WAN links, using legacy protocols such as MPLS. While MPLS is built to secure these p2p links between enterprise and data centers, it’s not designed to connect to the public cloud. The new trend if cloud and cloud connectivity: - The volume of enterprise traffic is growing rapidly across all global regions and verticals. - 50% of enterprises traffic is HTTP and HTTPS. Applications are moving from onpremises to the cloud. - Access-site BW is reasonably good worldwide. - Enterprises have the challenge of connecting long distance across the globe, having latency problems with TCP application response times, and with great variations with some geographies. - Enterprises try to keep up with the scale and dynamic nature of cloud connectivity, including connecting mobile users across the globe (dynamic SD-WAN, software and control provisioning) SD-WAN A Software-Defined Wide Area Network (SD-WAN) is a virtual WAN architecture that allows enterprises to leverage any combination of transport services (including MPLS, LTE and broadband internet services) to securely connect users to applications. MEF published in Ag. 2019 the SD-WAN Service Attributes and Services (MEF 70) standard. SD- WAN for 5G (Mapping SD-WAN application performance and security to 5G slices). INTRODUCTION TO TRANSPORT NETWORK 45 Key advantages include: - Reducing costs with transport independence across MPLS, 4G/5G LTE and other types of connections. Transmission is adapted to the most suitable service. - Reliability - Security - Improved OPEX, replacing MPLS services with more economical and flexible broadband (including secure VPN connections). - Improve application performance and increasing agility (dynamic route application traffic for efficient delivery and improved user experience). - Optimize user experience and efficiency for Software-as-a-Service(SaaS) and publiccloud applications. - Simplify operations with automation and cloud-based management. SD-WAN combines all data transport types into a single overlay connection that is intelligently managed to route network traffic over optimal physical network connections (fiber, DOCSIS, MPLS, 4G, 5G, etc.) SD-WAN can automate much of the complexity of tunnel monitoring, redundancy, and fail- over configurations. It is possible to maintain MPLS tunnels (Hybrid SD-WAN). STILL MPLS