Abstract: This article discusses the transmission of operational Ethernet services and the necessity of division. The article analyzes the requirements of operational Ethernet division and gives the division boundary through the main functional architecture. This article discusses some related technologies that divide the boundary, and compares and introduces the selection of different technologies.
A similar article was published in Maxim Engineering Journal, Issue 63 (PDF, 1MB).
The Metro Ethernet Forum (MEF) has developed many service standards for the promotion of Ethernet interconnection, collectively known as operational Ethernet services. The purpose of the MEF in formulating these services is to enhance the versatility of Ethernet by enhancing the interoperability of high-quality Ethernet services between various operators. To ensure that customers of these services can compare the quality of service providers, MEF has also developed measurable attributes for each service. These attributes are of great significance to the customer. Therefore, when the customer physically connects their network with the service provider's network, these attributes must be able to represent significant service characteristics. The physical interface of this connection is called the demarcation point.
The service division has been widely used in telephone service providers ("operators") for decades. The common division form is a small box installed outdoors. This small box connects the local operator's telephone network to the indoor distribution terminal, thereby providing wired telephone service to customers. This small box is the dividing line between the customer's responsibility and the operator's responsibility.
For telephone services, the functional requirements of the demarcation unit are the smallest, while the Ethernet service operator's functional requirements for the demarcation unit are much more complicated. Operators usually provide services to users in accordance with the service level agreement (SLA) contract, which specifies the service details, such as the promised information rate (CIR), throughput (CBS), available services, frame delay, Frame jitter, frame loss rate, and fault recovery time. The demarcation unit operating the Ethernet plays an important role in ensuring that the services actually provided comply with the SLA rules.
Demand of the demarcation unit From a functional point of view, the Ethernet demarcation unit must provide at least physical connections and measurement points: RJ-45 sockets or fiber optic connectors. IEEE® defines the interconnection of the physical layer, and the MEF technical specifications also refer to the content of the physical layer interconnection. The demarcation unit must accept the standard IEEE 802.3 Ethernet frame format from the user and prepare for transmission between the service provider networks. The minimum functional requirements depend on the specific application.
The user interface part of the demarcation function is called the user network interface (UNI). MEF stipulates the standard functional scope of UNI, from the most basic type 1 UNI to the automatically configurable type 3 UNI. Recently, MEF has passed the technical specification entitled MEF 13 UNI Type 1 ImplementaTIon Agreement and related test verification process. MEF is expected to further expand Type 2 and Type 3 UNI in MEF 11 in the near future.
Class 1.2 UNI of MEF 13 requires the demarcation unit to handle some of the work of the layer 2 protocol, which is used to establish the connection between the customer network and the UNI. This requirement makes the demarcation unit must have the operability and filtering capabilities of the layer 2 protocol. In addition, ITU and MEF require Type 2 and Type 3 UNIs to be able to implement certain management protocols for Layer 2 and require the demarcation unit to have comprehensive Layer 2 protocol processing capabilities—at least at minimum throughput. These necessary conditions affect the technical decision of the demarcation unit design. A favorable factor for these necessary conditions is that most boundary cell designs can increase the processing power of valuable high-level application functions.
Structure of the demarcation unit Figure 1 is a basic block diagram of the Ethernet demarcation unit. The "UNI" on the right side of the figure is the schematic diagram of the user network interface, and the "Network interface" on the left side is the schematic diagram of the operation transmission network interface. The network interface function group is called a network interface device (NID). NID represents an independent part of the device, but the NID here is used to illustrate a part of the function of the demarcation unit. The MEF has announced the draft of the NID functional standard, but it has not been officially approved yet. The NID function may or may not represent the same device as the UNI. In practical applications, the dividing line between UNI and NID varies with different applications. To further clarify the issue, MEF has developed a chart that divides the scope of UNI. It should be noted that the demarcation unit usually includes a UNI network (UNI-N) interface or an interface between external networks (E-NNI).
Figure 1. Functional block diagram of the Ethernet demarcation unit. UNI includes the functional block diagram of the user network interface, and below the NID is the functional block diagram of the interface that operates the transmission network.
The hardware of the UNI and NID data platforms is closely related to the corresponding network technology. The physical interface of the UNI must include an electrical or optical Ethernet interface that conforms to the IEEE standard, and the operating optical network connected to the NID is usually SONET / SDH. These different technologies need to be converted to realize the intercommunication of user data between the various operator networks, and the two data platforms need to have conversion and interworking functions. The interoperability of the data platform is achieved through an integrated Ethernet mapper, a network processing unit (NPU), or a custom-made field programmable gate array (FPGA). In the case of SONET / SDH, the integrated Ethernet-over-SONET / SDH (EoS) mapper is usually the best solution to realize the interoperability of data platforms. The data platform is used for the transmission and marking of customer data and flow control or data shaping.
UNI and NID usually have independent control platform software that can handle the underlying configuration and status monitoring. The following are examples of effective processing through the control platform: detection of cable connection, status and performance monitoring, and handling of interrupt events. The control platform ensures that services are provided in accordance with the instructions of the management platform. The control platform is usually implemented by software running on a local microprocessor that configures and controls the hardware of the data platform. In independent demarcation units, the control platforms of UNI and NID usually run on the same processor. The control platform and data platform can sometimes be implemented on an NPU, although this will cause conflicts between complex data and control platforms. It is easier to implement an architecture with a separate data platform and control platform.
The management platform is usually implemented by software on line cards or racks. Both UNI and NID have management platforms for fault recovery and SLA performance monitoring. The main role of the management platform is to deal with issues related to network operations. MEF has defined the architecture of network operation management information in MEF 7 EMS-NMS InformaTIon Model. In addition, in addition to the network-oriented management platform, UNI also has a user-oriented management platform. MEF defines the basic architecture of the user-oriented management interface in MEF 16 Ethernet Local Management Interface (E-LMI). The management platform of UNI and NID can exchange information, but this is not necessary.
The UNI management platform also uses the Ethernet management protocol developed by ITU-T and IEEE, called Ethernet OAM (Operation, Management and Maintenance). IEEE 802.3ah OAM is used to monitor the operating status of a point-to-point link and improve fault isolation. ITU-T Y.1731 OAM expands the range of functions to include advanced multi-link operations such as link tracking, connectivity detection, and automatic protection switching. With OAM, network operators can perform management tasks that were difficult for Ethernet a few years ago.
UNI technology UNI network transmission technology is limited to the full-duplex 10Mbps, 100Mbps, 1Gbps MEF standard or 10Gbps IEEE compatible electrical or fiber optic Ethernet standard. In order to effectively cover the scope of physical layer connections, some demarcation units are only allowed to insert small pluggable (SFP) modules in order to configure the physical interface of the UNI during installation. Use Ethernet mapper, NPU or FPGA to achieve the required functions of other UNI data platforms, regardless of the physical interface. Another alternative to SFP is to use on-board PHY or fiber optic modules. This alternative solution achieves lower design costs at the expense of flexibility.
NID technology driven by network positioning Figure 2 shows two common demarcation environments. In Site A (Surrounding Boundary), the building where the user is located can be directly connected to the optical fiber network, or very close to the optical fiber network. This situation is common in metropolitan area networks and automatic switched optical networks called E-NNI. When the demarcation point is located at the border of the optical fiber network, the NID must be directly connected to the optical fiber network or connected to the low-rate data branch accessed through the optical network device. For the application of optical fiber network boundaries, the selection of NID technology involves multiple fields: traditional SONET / SDH, Ethernet-over-PDH-over-SONET / SDH (EoPoS), operator's backbone network bridging (PBB), transmission multi-protocol labeling Switching (T-MPLS), Passive Optical Network (PON), Dense Wavelength Division Multiplexing (DWDM), Resilient Packet Ring (RPR), Fiber Ethernet and Hybrid Fiber Coaxial Cable (HFC), etc.
Figure 2. Two common demarcation sites. Venue A is commonly found in a metropolitan area network. The building where the user is located can be directly connected to the optical fiber network or close to the optical fiber network. Site B is commonly found in remote areas, and it is necessary to extend the connection of the fiber network to the building where the user is located through the "last mile" technology.
In the place B (under-circumferential boundary) in Figure 2, the optical fiber network does not reach the building where the user is located, and must use the "last mile" technology to reach the boundary point. This kind of situation is common in cities, suburbs and rural areas with low population density. For the case where the demarcation point is far from the boundary of the optical fiber network, NID technology includes: Ethernet-over-PDH (EoPDH), digital subscriber link (DSL), Miles (EFM) and Wired Cable Data Service (DOCSIS).
Which transmission technology is used for a specific NID depends on the cost-effectiveness of the interface with the existing network of the operator providing the service. The architecture of the demarcation platform is preferably a modular design to cover the required UNI and NID transmission technologies.
Technical development level of Ethernet mapping Generally speaking, in the Ethernet demarcation unit, the integrated Ethernet mapper can realize the UNI-NID interworking function. EoS and EoPDH mapping are the two most common mapping technologies. EoS mapping has been used for nearly ten years, while EoPDH mapping has been used for less than two years. Maxim's recently launched EoPDH Ethernet mapper DS33X162 series can fully meet the current requirements for demarcation design. This series of products includes nine devices and supports one to sixteen PDH links. It is the industry's only fully support ITU standard EoPDH mapping, using Device with unified software design / hardware pinout. All devices in this series adopt strict separation of control platform and data platform, and can flexibly execute high-level protocols.
Expansion of minimum requirements At present, MEF only stipulates the minimum requirements of UNI functions, and equipment manufacturers have enough space to integrate value-added functions, so that the demarcation unit has greater competitiveness while maintaining compatibility. Value-added functions are divided into two categories: one is a function that facilitates the sale of demarcation units to operators, and the other is a function that facilitates operators to sell services to users.
Installation cost is the first concern of operators. The demarcation unit with automatic presetting and internal network diagnosis function can greatly reduce the overall cost. Using these technologies can support medium-bandwidth Ethernet, and can reuse existing equipment and infrastructure, such as EoPDH, thereby reducing operator costs. Real-time SLA monitoring and seamless integration with the operator's existing network management system (NMS) also significantly improve the operator's value-added services.
In addition, providing users with convenient and reliable services is also a key evaluation criterion. For example, the management interface based on the website browser (HTML) can easily observe the UNI running status of all users, which is a very convenient function. This interface can be provided by the UNI itself or integrated in the central office to collect information from all users UNI. A similar HTML user interface can provide SLA-compliant reports, bandwidth utilization status, and quality of service configuration options for each UNI at different times.
Users can also use VLAN tags to provide different quality services for different VLANs or applications. Enterprise network managers need to use their existing network management tools (such as SNMP) to automatically monitor the status, or receive alarm notifications via email. As the volume of business increases, growing companies hope to continue to expand their bandwidth. NID technology uses VCAT / LCAS link set, easy to expand bandwidth. The demarcation unit can provide dynamic presets, increase bandwidth when demand is high, and reduce bandwidth when demand is low. These are just a few examples, but provide some suggestions for the future direction of development.
At present, MEF has developed a public MEF 13 platform, and the demarcation unit architecture adopts a common infrastructure, on which the future demarcation technology is built.
A similar article was published in Lightwave in May 2008.
IEEE is a registered service mark of the American Institute of Electrical and Electronics Engineers.
