A comprehensive explanation of cell Packaging Technology in Cisco Routers

It is very useful to learn about the cell packaging technology. So I studied the cell Packaging Technology in Cisco routers. I will share it with you here, hoping it will be useful to you. Cell packaging, also known as cell concatenation) is a mechanism that encapsulates multiple cell relay asynchronous transmission modes (ATM) cells into the same IP/MPLS group. It enables telecom operators to overcome the low bandwidth efficiency inherent in cell relay transmission.

Low cell relay bandwidth efficiency

The reason why cell relay causes low bandwidth efficiency of the group switching network PSN is mainly due to the following reasons. Cell relay requires switching all cell relay groups, which reduces the packet forwarding rate (PPS ). For example, if six ATM cells need to be transmitted, the network core device must exchange six groups, consuming resources such as bandwidth forwarded by six groups ). However, if the cell packaging technology is used to package six ATM cells into the same group, the core network device only needs to exchange one group and only needs to consume the resources forwarded by one group. Therefore, telecom operators can use the cell packaging technology to save core network group forwarding resources.

Cell relay also reduces bandwidth utilization. For example, assume that a telecom operator wants to transmit a cell to relay an ATM Cell stream on a Multi-Protocol Label exchange MPLS core that includes the packet-over-SONETPOS interface. On the edge of the carrier network, 52-byte ATM cells do not include one byte header verification [HEC]). comes with Layer 2 headers that contain 4-byte control words, 4-byte virtual circuit tags, 4-byte tunnel tags, and 4-byte Cisco Advanced Data Link Control HDLC. The total overhead of each 52-byte ATM cell is 16 bytes. Therefore, the final group size on the POS optical fiber is 68 bytes, that is, 52 bytes of ATM cells + 16 bytes of overhead ). In this case, the bandwidth waste rate of the cell relay is about 23.52% 16/68 ).

In this case, if the cell packaging technology is used and six cells are encapsulated in the same MPLS group, the total overhead is still 16 bytes, however, the payload is 312 bytes, which increases the bandwidth efficiency to about 95.12%. Therefore, telecom operators can use credit packaging technology to increase bandwidth utilization by 18.65%, that is, from 76.47% to 95.12% ). Cisco 12000, 7500, and 7200 vrouters support cell packaging technology, but this article mainly introduces it in the 4-port IP service engine ISE) implementation of ATM-over-SONET OC-12/STM-4 and 4-port ise atm-over-SONET OC-3/STM-1 line card.
 
Bandwidth saving

ATM cells are packaged into an MPLS Group to improve the efficiency of the group Switching Network (PSN.

Cell packaging Parameters

The cell packaging technology mainly includes two parameters: the minimum number of cells to package MNCP) size is also known as the cell packaging size) and the maximum cell packaging timer MCPT) Timeout value. When establishing a technical connection for cell packaging, you must configure the MNCP size and MCPT timeout value. This information is recorded in the input and output hardware cell packaging Field Programmable Gate Array FPGA. The range of these parameters is as follows:
◆ MNCP is between 2 and 28 ATM cells.
◆ The MCPT timeout value ranges from 2 to 4095 microsecond us)
◆ The precision of MCPT hardware programming ranges from 50us to 25msMCPT timeout is 50us)

The maximum value of MNCP is 28 ATM cells, because the MTU of Ethernet is 1500 bytes. If an MPLS group contains more than 28 ATM cells, it is discarded on the Ethernet interface. Although the Cisco IOS command line interface CLI) allows you to set the MCPT value to a value between 2-40 95us, the hardware programmable range is actually 50 us to 25 ms, the step unit is 50 US.

During tag propagation and binding, PE1 and PE2 exchange MNCP values by setting the LDP interface parameter fields. When PE1 receives the MNCP of PE2, it is stored in the pre-defined VC/VP/Port Database of PE1, and vice versa. Any change to MNCP on any vendor edge will result in the Undo of the tag and re-establish the virtual circuit at both ends, and the old value will be replaced by the new value.

If PE1 does not support cell packaging, that is, MNCP is equal to 1), PE2 should send only one cell in each MPLS group, but be able to receive packaged cells-If PE2 starts the cell packaging technology. MCPT plays an important role locally, and its range usually depends on the ATM connection speed OC-3 or OC-12. If the MCPT timer times out, the packaged cells are immediately sent through an MPLS group-even if the package is not completed, the number of cells in the group does not reach MNCP. The lifecycle of a cell Packaging Technology Group in an MPLS network.

At the edge of the input supplier, ATM cells reach the ATM port and are processed by segmentation and restructuring SAR chips. The SAR chip divides each ATM Cell into AAL0, AAL5, or OAM cells. For valid AAL0 cells, the SAR chip removes a HEC byte from the ATM cell header and sends the remaining 52-byte ATM cells to the cell packaging FPGA. The FPGA package the cells according to the MNCP and MCPT configuration parameters. Input ATM cells are queued up to form a cell package until the predefined cell package size is reached. In this case, FPGA generates a control word, encodes the T sign into 0, specifies the AAL0 ATM cell type, and generates a serial number for each cell package group. All the remaining fields are set to 0.