No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

OSN 500 550 580 V100R008C50 Commissioning and Configuration Guide 02

Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Basic Concepts

Basic Concepts

Before you configure Ethernet boards with services, you need to learn the basic concepts including external port, internal port, logical port, and bridge so that you can understand the service configuration process and the signal flow when the boards process the services.

Formats of Ethernet Frames

To implement the VLAN and QinQ functions, the IEEE 802.1q and IEEE 802.1ad protocols define different formats of the Ethernet frames, which contain different VLAN information.

To implement the VLAN function, the IEEE 802.1q protocol defines the Ethernet frame format that contains the VLAN information. Compared with the ordinary Ethernet frame, the frame with the format defined by the IEEE 802.1q protocol is added with a four-byte header.

To implement VLAN mesting (QinQ), the IEEE 802.1ad protocol defines two VLAN tag types. See Figure 2-52. The VLAN tag types are defined to differentiate the services on the client side and the services on the supplier service side.

  • The VLAN tag used on the client side is represented as C-VLAN, of which the frame format is the same as the frame format defined by the IEEE 802.1q protocol.
  • The VLAN tag used on the supplier service side is represented as S-VLAN.
Figure 2-52  Formats of Ethernet frames

The length of the data field is variable. maximum length of the data field depends on the maximum frame length that the ports of the equipment support.

The four-byte S-VLAN or C-VLAN field is divided into two sub-fields: the tag protocol ID (TPID) and the tag control Information (TCI).

Both the TPID and TCI consist of two bytes. See Figure 2-53.

Figure 2-53  Positions of the TPID and TCI in the frame structure

  • TPID structure

The TPID consists of two bytes and indicates the VLAN tag type. TPID of the C-VLAN is always 0x8100 whereas the TPID of the S-VLAN can be customized. Refer to Table 2-148.

Table 2-148  Tag types defined by using the TPID
Tag Type Name Value
C-VLAN Tag 802.1q Tag Protocol Type 0x8100
S-VLAN Tag 802.1q Service Tag Type Customizable
NOTE:

The IEEE 802.1ad specifies the TPID of the S-VLAN to 0x88a8. In actual application, the setting of TPID for the S-VLAN tag varies according to the equipment manufacturer. To ensure compatibility between interconnected equipment, it is recommended that you set the TPIDs of the S-VLAN tags of the interconnected equipment to the same value within 0X600-FFFF.

  • TCI structure

The TCI structure of the S-TAG is basically the same as the TCI structure of the C-TAG. VLAN ID (VID) field consists of 12 bits and ranges from 0 to 4095. The difference is that the TCI of the S-TAG contains the drop eligible (DE) indication and works with the priority code point (PCP) to indicate the priority of the S-TAG frame.

The TCI structures of the C-TAG and S-TAG are shown in Figure 2-54 and Figure 2-55.

Figure 2-54  TCI structure of the C-TAG

The TCI field of the C-TAG consists of the following bytes:

  • PCP: three bits
  • CFI: one bit
Figure 2-55  TCI structure of the S-TAG

The TCI field of the S-TAG consists of the following bytes:

  • PCP: three bits
  • DE: one bit
  • VID: 12 bits

Internal Ports and External Ports

External ports on Ethernet boards are used to access the services on the user side. Internal ports on Ethernet boards are used to encapsulate and map the services to the transmission network for transparent transmission.

External ports on Ethernet boards (that is, external physical ports) are also referred to as client-side ports or user-side ports, which are used to access the Ethernet services on the user side.

Internal ports on Ethernet boards (that is, internal VCTRUNKs) are also referred to as system-side ports or backplane-side ports in certain cases, which are used to encapsulate and map the services to the SDH side.

VCTRUNKs are VC-based transmission paths, which can be implemented by using the adjacent concatenation or virtual concatenation technology. On the U2000 window, paths are bound to specify the bandwidth of different granularities for a VCTRUNK port.

Figure 2-56  External ports and internal ports on Ethernet boards

Auto-Negotiation

The auto-negotiation function allows the network equipment to send information of its supported working mode to the opposite end on the network and to receive the corresponding information that the opposite end may transfer.

The working modes of the interconnected ports on the equipment at both ends must be the same. Otherwise, the services are affected.

If the working mode of the port on the opposite equipment is full duplex and if the working mode of the port on the local equipment is auto-negotiation, the local equipment works in the half-duplex mode. That is, the working modes of the interconnected ports at both ends are different, and thus packets may be lost. Hence, when the working mode of the port on the opposite equipment is full duplex, you need to set working mode of the port on the local equipment to full duplex.

NOTE:

When the interconnected ports at both sides work in the auto-negotiation mode, the equipment at both sides can negotiate the flow control through the auto-negotiation function.

The auto-negotiation function uses fast link pulses (FLPs) and normal link pulses (NLPs) to transfer information of the working mode so that no packet or upper layer protocol overhead needs to be added.

NOTE:

This topic considers FE electrical interfaces as an example to describe how to implement the auto-negotiation function.

The FLP is called the 100BASE-T link integrity test pulse sequence. Each set of equipment on the network must be capable of issuing FLP bursts in the case of power-on, issuing of management commands, or user interaction. The FLP burst consists of a series of link integrity test pulses that form an alternating clock/data sequence. Extraction of the data bits from the FLP burst yields a link code word that identifies the working modes supported by the remote equipment and certain information used for the negotiation and handshake mechanism.

To maintain interoperability with the existing 100BASE-T equipment, the auto-negotiation function also supports the reception of 100BASE-T compliant link integrity test pulses. The 10BASE-T link pulse activity is referred to as the NLP sequence. equipment that fails to respond to the FLP burst sequence by returning only the NLP sequence is treated as the 100BASE-T compatible equipment.

The first pulse in an FLP burst is defined as a clock pulse. Clock pulses within an FLP burst occur at intervals of 125 us. Data pulses occur in the middle of two adjacent clock pulses. The positive pulse represents logic "1" and the absence of a pulse represents logic "0". An FLP burst consists of 17 clock pulses and 16 data pulses (if all data bits are 1). The NLP waveform is simpler than the FLP waveform. NLP sends a positive pulse every 16 ms when no data frame needs to be transmitted.

Figure 2-57  Waveform of a single FLP

Figure 2-58  Consecutive FLP and NLP bursts

Flow Control

When the data processing/transferring capability of the equipment fails to handle the flow received at the port, congestion occurs on the line. To reduce the number of discarded packets due to buffer overflowing, proper flow control measures must be taken.

The half-duplex Ethernet port applies the back-pressure mechanism to control the flow. The full-duplex Ethernet port applies PAUSE frames to control the flow. Currently, the half-duplex Ethernet function is not widely applied. Hence, the flow control function realized by Ethernet service boards is used for the full-duplex Ethernet ports.

The flow control function realized by Ethernet service boards is classified into two types: auto-negotiation flow control and non-auto-negotiation flow control.

Auto-Negotiation Flow Control
When the Ethernet port works in the auto-negotiation mode, you can adopt the auto-negotiation flow control function. The auto-negotiation flow control modes include the following:
  • Enable dissymmetric flow control

    The port can transmit PAUSE frames in the case of congestion but cannot process the received PAUSE frames.

  • Enable symmetric flow control

    The port can transmit PAUSE frames and process the received PAUSE frames.

  • Enable symmetric/dissymmetric flow control

    The port has the following abilities:

    • Transmits and processes PAUSE frames.
    • Transmits PAUSE frames but cannot process the received PAUSE frames.
    • Processes the received PAUSE frames but cannot transmit PAUSE frames.
  • Disable

    Disables the auto-negotiation flow control function.

Non-Auto-Negotiation Flow Control
When the Ethernet port works in a fixed working mode, you can adopt the non-auto-negotiation flow control function. The non-auto-negotiation flow control modes include the following:
  • Send only

    The port can transmit PAUSE frames in the case of congestion but cannot process the received PAUSE frames.

  • Receive only

    The port can process the received PAUSE frames but cannot transmit PAUSE frames in the case of congestion.

  • Send and receive

    The port can transmit PAUSE frames and process the received PAUSE frames.

  • Disable

    The port does not support the auto-negotiation flow control function.

Realization Principle

The realization principle of the flow control function is described as follows:

  1. When congestion occurs in the receive queue of an Ethernet port (the data in the receive buffer exceeding a certain threshold) and the port is capable of sending PAUSE frames, the port sends a PAUSE frame to the peer end. The pause-time value in the frame is N (0<N≤65535). pause-time indicates the time interval between two packets sent from the peer end.
  2. If the Ethernet port at the opposite end is capable of processing PAUSE frames, this Ethernet port stops sending data within a specified period of time N (the unit is the time needed for sending 521 bits) after receiving the PAUSE frame.
  3. If the congestion at the receive port is cleared and the pause-time at the peer port ends, the peer port starts to send data to the receive port. If the congestion at the receive port is not cleared, the port sends a PAUSE frame whose pause-time is not 0 to extend the pause time. If the congestion at the receive port is cleared (the data in the receive buffer is below a certain threshold) but the pause-time does not end, the port sends a PAUSE frame whose pause-time is 0 to notify the opposite end to send data.

IEEE 802.3 defines the format of the PAUSE frame as follows:

  • Destination address: 01-80-C2-00-00-01 (multicast address)
  • Source address: MAC address of the source port
  • Type/Length: 88-08 (MAC control frame)
  • MAC control code: 00-01 (PAUSE frame)
  • MAC control parameter: pause-time (two bytes)
Figure 2-59  Structure of the PAUSE frame

Encapsulation and Mapping Protocol

To ensure that Ethernet frames can be transparently transmitted over the optical transmission network, the Ethernet frames need to be encapsulated and mapped into VC containers at the access point. The encapsulation and mapping protocols used by the Ethernet service board include the high-level data link control (HDLC), link access procedure - SDH (LAPS), and generic framing procedure (GFP).

HDLC

The HDLC is a general data link control procedure. When using the HDLC protocol, the system encapsulates data services into HDLC-like frames as information bits and maps the frames into SDH VC containers.

LAPS

The LAPS is also a data link control procedure. It is optimized based on the HDLC. The LAPS complies with ITU-T X.86.

GFP

The GFP is the most widely applied general encapsulation and mapping protocol. It provides a general mechanism to adapt higher-layer client signal flows into the transport network and can map the variable-length payload into the byte-synchronized transport path. The client signals can be protocol data units (PDU-oriented, such as IP/PPP and Ethernet), block code data (block-code oriented, such as Fiber Channel and ESCON), or common bit data streams. The GFP protocol complies with ITU-T G.7041.

The GFP defines the following modes to adapt client signals:

  • Frame-mapped GFP (GFP-F)

    The GFP-F is a PDU-oriented processing mode. It encapsulates the entire PDU into the GFP payload area and makes no modification on the encapsulated data. It determines whether to add a detection area for the payload area, depending on requirements.

  • Transparent GFP (GFG-T)

    The GFP-T is a block-code (8B/10B code block) oriented processing mode. It extracts a single character from the received data block and maps the character into the fixed-length GFP frame.

Virtual Concatenation

The rate of the Ethernet service does not adapt to the rate of the standard VC container. Hence, if you directly map the Ethernet service data into a standard VC container, there is a great waste of the transmission bandwidth. To solve the problem, use the virtual concatenation technology to concatenate many standard VC containers to a large VC container that adapts to the rate of the Ethernet service.

The concatenation is defined in ITU-T G.707 and contains contiguous concatenation and virtual concatenation. Both concatenation methods provide concatenated bandwidth of X times Container-N at the path termination.

Contiguous concatenation concatenates the contiguous VC-4s in the same STM-N into an entire structure to transport. It maintains the contiguous bandwidth throughout the whole transport. Virtual concatenation concatenates many individual VC containers (VC-12 containers, VC-3 containers, or VC-4 containers) into a bit virtual structure to transport. The virtual concatenation breaks the contiguous bandwidth into individual VCs, transports the individual VCs, and recombines these VCs to a contiguous bandwidth at the transmission termination point.

In the case of virtual concatenation, transport of each VC container may occupy different paths and there may be a transport delay difference between VC containers. Hence, there are difficulties to restore the client signal. Virtual concatenation requires concatenation functionality only at the path termination equipment and it can flexibly allocate bandwidth. Hence, the virtual concatenation technology is widely used.

Virtual concatenation is available in two types: virtual concatenation in a higher order path and virtual concatenation in a lower order path. A higher order virtual concatenation VC4-Xv provides a payload of X Container-4s (VC-4s). The payload is mapped individually into X independent VC-4s. Each VC-4 has its own POH. A lower order virtual concatenation VC-12-Xv provides a payload of X Container-12s (VC-12s). The payload is mapped individually into X independent VC-12s. Each VC-12 has its own POH. It is the same case with the virtual concatenation of VC-3s.

VC4-Xv and VC-3-Xv

The virtual container that is formed by a VC4-Xv/VC-3-Xv can be mapped into X individual VC-4/VC-3s that form the VC4-Xv/VC-3-Xv. Each VC-4/VC-3 has its own POH. POH has the same specifications as the ordinary VC-4 POH. The H4 byte in the POH is used for the virtual concatenation-specific multiframe indicator (MFI) and sequence indicator (SQ).

MFI indicates the position of a frame in the multiframe. Each frame sent by the source carries the MFI information. The sink end combines the frames with the same MFI into the C-n-Xv. MFI includes MFI-1 and MFI-2. MFI-1 is transmitted by bits 5-8 of the H4 byte and ranges from 0 to 15. MFI-2 is transmitted by the two frames of which the MFI-1 is "0" and "1" in the multiframe. Bits 1-4 of the H4 bytes of the two frames indicate the higher four bits and lower four bits of the MFI-2 respectively. Hence, the MFI-2 ranges from 0 to 255. That is, a multiframe consists of 4096 frames and the period is 512 ms.

SQ indicates the position of a frame in the C-n-Xv. The source end inserts the SQ information into the frame according to the payload allocation sequence. The sink end determines the sequence to extract the payload from the frames that form C-n-Xv according to the SQ. SQ is transmitted by the two frames of which the MFI-1 is "14" and "15" in the multiframe. Bits 1-4 of the H4 bytes of the two frames indicate the higher four bits and lower four bits of the SQ respectively.

Figure 2-60  VC-3-Xv/VC4-Xv multiframe and sequence indicator

With the MFI and SQ, the sink end can correctly restore the position of each frame in the C-n-Xv to prevent the frame alignment problem due to the different propagation delays of the frames.

VC-12-Xv

The virtual container that is formed by a VC-12-Xv can be mapped into X individual VC-12s which form the VC-12-Xv. Each VC-12 has its own POH. POH has the same specifications as the ordinary VC-12 POH. Bit 2 of the K4 byte in the POH is used for the virtual concatenation-specific frame count and sequence indicator.

Bit2s of the K4 bytes in every 32 multiframes (one multiframe comprising four VC-12s) are extracted to form a 32-bit character string to express the frame count and sequence indicator. Bits 1-5 of the string express the frame count, whose value range is between 0 and 31. structure formed by 32 multiframes has 128 frames. Hence, the resulting overall multiframe is 4096 frames with the period of 512 ms. Bits 6-11 of the string express the sequence indicator. The frame count/sequence indicator in the VC-12-Xv has the same usage as the multiframe indicator/sequence indicator in the VC4-Xv/VC-3-Xv.

Tag Attribute

When data frames are received on or transmitted from a port on an Ethernet board, the processing mode of the data frames is determined by the tag attributes of the port.

The tags for the port on an Ethernet board are available in three types: tag aware, access, and hybrid.

Table 2-149  Processing mode of data frames on ports with different tags
Direction Data Frame Type Processing Mode
Tag aware Access Hybrid
Ingress port Data frames with VLAN tags The data frames are transparently transmitted. The data frames are discarded. The data frames are transparently transmitted.
Data frames without VLAN tags The data frames are discarded. The VLAN tags that contain Default VLAN ID and VLAN Priority are added to the data frames, and then the data frames are transparently transmitted.
Egress port Data frames with VLAN tags The data frames are transparently transmitted. After the VLAN tags are stripped from the data frames, the data frames are transparently transmitted.
  • If the VLAN IDs contained in the data frames are Default VLAN ID, the VLAN tags are stripped from the data frames, and then the data frames are transparently transmitted.
  • If the VLAN IDs contained in the data frames are not Default VLAN ID, the data frames are transparently transmitted.
NOTE:

The tag setting is valid only when the following conditions are met:

  • Port Type of the port is set to PE or UNI.
  • The entry detection function is enabled. When the Ethernet switching board works in the Ethernet transparent transmission state and when the entry detection function is disabled, the port transparently transmits the received data frames regardless of whether the data frames carry VLAN tags.

Based on the features of tag aware, access, and hybrid, adhere to the following principles when setting the tag for a port:

  • If the data packets transmitted from the interconnected equipment carry VLAN tags, set TAG to Tag Aware.
  • If the data packets transmitted from the interconnected equipment do not carry VLAN tags, set TAG to Access.
  • If the data packets transmitted from the interconnected equipment may carry VLAN tags, set TAG to Hybrid.

Bridge

A bridge is a functional unit that is used to implement the interconnection between two or more LANs.

VB and LP

The switching domain of an Ethernet board that has the Layer 2 switching capability can be divided into multiple sub-switching domains. As a result, if no services are interconnected, different various bridges (VBs) cannot access each other. Each VB has an independent configuration mode and uses an independent VLAN. Different VBs can use the same VLAN.

A VB can contain a number of logical ports (LPs). By configuring the mounting relation, you can mount multiple PORTs and VCTRUNKs to the same VB.

Figure 2-61 shows the relations between VBs, LPs, PORTs, and VCTRUNKs.

Figure 2-61  Relations between VBs, LPs, PORTs, and VCTRUNKs

Transparent Bridge and Virtual Bridge
  • The services of different transparent bridges are isolated, but the services in the same transparent bridge are not isolated. The entire transparent bridge is a switching domain.
  • The services of different virtual bridges are isolated and the services with different VLAN IDs in the same virtual bridge are also isolated. The switching domain of the entire virtual bridge is divided into multiple sub-switching domains according to the VLAN IDs.
NOTE:

As shown in Figure 2-62, the same logical port may belong to one or more sub-switching domains with different VLAN IDs. On the U2000, the same logical port can belong to multiple filtering tables with different VLAN IDs.

Figure 2-62  Transparent bridge and virtual bridge

Table 2-150  Transparent Bridge and Virtual Bridge
Item Transparent Bridge Virtual Bridge
VLAN filtering table It is not configured. It must be configured.
Ingress filter Does not check the validity of VLAN tags. All data frames that enter the bridge are valid. Check the validity of VLAN tags. If the VLAN ID is not the same as the VLAN ID defined in the VLAN filtering table, discard the data frame.
MAC address learning mode SVL IVL
Data frame forwarding mode Query the MAC address table to obtain the forwarding port of the data frame according to the destination MAC address of the data frame. Query the MAC address table to obtain the forwarding port of the data frame according to the destination MAC address and VLAN ID of the data frame.
Broadcast range Forward broadcast data frames on all ports of a bridge. Forward the broadcast data frames on the forwarding ports defined in the VLAN filtering table.
NOTE:

To forward a Layer 2 switching service, a bridge must learn the MAC address. A bridge learns the MAC address through one of the following methods: shared VLAN learning (SVL) and independent VLAN learning (IVL).

  • When the bridge adopts the SVL learning mode, the entry in the MAC address table is created according to the source MAC address and source port of the data frame. The entry is valid for all VLANs.
  • When the bridge adopts the IVL learning mode, the entry in the MAC address table is created according to the source MAC address, VLAN ID, and source port of the data frame. The entry is valid for only the VLAN.
Bridge Type

As listed in Table 2-151, the Ethernet boards support three types of bridges.

Table 2-151  Types of bridges supported by the Ethernet boards
Bridge Type Bridge Switch Mode Bridge Learning Mode

IEEE 802.1d MAC bridge

SVL/Ingress Filter Disable SVL

IEEE 802.1q Virtual Bridge

IVL/Ingress Filter Enable IVL

IEEE 802.1ad Provider Bridge

1 SVL/Ingress Filter Disable SVL
2 IVL/Ingress Filter Enable IVL
  • IEEE 802.1d MAC bridge: The IEEE 802.1d MAC bridge does not check the contents of the VLAN tags that are in the data frames, but performs Layer 2 switching according to the destination MAC addresses of the data frames.
  • IEEE 802.1q bridge: The IEEE 802.1q bridge supports data isolation by using one layer of VLAN tags. This bridge checks the contents of the VLAN tags that are in the data frames and performs Layer 2 switching according to the destination MAC addresses and VLAN IDs.
  • The IEEE 802.1ad bridge: The IEEE 802.1ad bridge supports data frames with two layers of VLAN tags. This bridge adopts the outer S-VLAN tags to isolate different VLANs and supports only the mounted ports whose attributes are C-Aware or S-Aware. This bridge supports the following switching modes:

    1. This bridge does not check the contents of the VLAN tags that are in the data frames, but performs Layer 2 switching according to the destination MAC addresses of the data frames.
    2. This bridge checks the contents of the VLAN tags that are in the data frames and performs Layer 2 switching according to the destination MAC addresses and the S-VLAN IDs of the data frames.
MAC Address Table

The entries in the MAC address table reflect the corresponding relations between MAC addresses and ports. The MAC address table contains the following entries:

  • Dynamic entry

    Indicates the entry that the bridge obtains by adopting the SVL/IVL learning mode. The dynamic entry ages and is even lost after the Ethernet switching board is reset.

  • Static entry

    Indicates the entry corresponding to the MAC address and the port that the network administrator manually adds in the MAC address table on the U2000. A static entry is a unicast entry. The static entry does not age and is not lost after the Ethernet switching board is reset.

  • Blackhole entry

    Indicates the entry used to discard the data frame that contains the specified destination MAC address, and is also referred to as the MAC address disable entry. The blackhole entry is configured by the network administrator. This entry does not age and is not lost after the Ethernet switching board is reset.

NOTE:
  • If a MAC address table is not updated within a specific period of time, that is, if the MAC address fails to be learnt because the new data frame from the MAC address is not received, this MAC address table is deleted automatically. This mechanism is considered as aging, and this period of time is considered as the aging time. The aging time of the MAC address table is five minutes by default and can be set by using the U2000.
  • A limited number of MAC addresses can be learnt at a time.
Hub/Spoke

Generally, the central station and non-central stations can access each other, but the non-central stations cannot access each other in the case of convergence services. Therefore, the ports mounted to the bridge need to be defined as Hub or Spoke ports.

  • Hub port

    Hub ports can access each other. Hub ports and Spoke ports can also access each other.

  • Spoke port

    Spoke ports cannot access each other. Hub ports and Spoke ports can access each other.

The mounted ports are Hub ports by default.

Translation
Download
Updated: 2019-01-21

Document ID: EDOC1100020976

Views: 45716

Downloads: 381

Average rating:
This Document Applies to these Products
Related Documents
Related Version
Share
Previous Next