JP2968677B2 - Address translation method for ATM networks - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an address conversion method for an ATM network, and more particularly, to an address conversion method for converting addresses used in a front-end network and VPI and VCI used in an ATM network. The present invention relates to an address conversion method for an ATM network that learns dynamically.

[0002]

2. Description of the Related Art Conventionally, as an address conversion system of this type, there is an automatic address learning function provided in an Internet device such as a MAC bridge. This function is limited to the case where the data packets used between the connected networks are the same (or the addressing method is the same). It is also used for packet filtering and forwarding of data. On the other hand, in the field of the ATM network, a conversion method relating to switching or routing of fixed-length packets (hereinafter, referred to as "cells") is disclosed in, for example, Japanese Patent Laid-Open No. 19452/1991.
It is assumed that a data packet to be switched or routed is a cell on both the input side and the output side.

[0003]

When an internetwork using an ATM network as a backbone is to be constructed, if an addressing method is different between the front-end network and the backbone network, an address used in the front-end network and an ATM are used. Virtual path identifier (V
PI) and a virtual channel identifier (VCI). At this time, there is a method of managing the address translation information manually, but the address translation information is manually registered and updated every time a data terminal is moved, added, or changed which is practically necessary in network management. Then, as the size of the network increases, the effort of the system administrator for the change work becomes enormous, which may become impossible.

[0004]

According to the address conversion method of the present invention, a virtual path identifier VPI field and a virtual channel identifier VCI field in a fixed length packet cell including a header part and an information part in a header are respectively defined as a destination field and a virtual field identifier. By dividing into the source field, the data terminal address of the source front-end network is extracted from the cell input from the ATM network side, and is associated with the source VPI and VCI in the header of the cell. Address learning means for storing address translation information, search means for searching for the address translation information by referring to the destination address in the cell input from the front-end network, and when already learned according to the search result Generate an individual destination cell If the learning has a cell generating means for generating unlearned cell.

[0005]

Next, the present invention will be described with reference to the drawings.

FIG. 1 is a network configuration diagram showing an embodiment of the present invention. Referring to FIG. 1, an embodiment includes an ATM node 200 that accommodates a front-end network (hereinafter, referred to as “FLAN”), and an ATM switching device 1 that switches cells from each ATM node 200.
00. In the ATM node 200, F
Address translation device 400 that translates data packets sent from the LAN into cells or vice versa
And a multiplexer 300 for multiplexing cells from each address translator 400 or demultiplexing cells from the ATM switch 100. The VPI (virtual path identifier) functions as an address for identifying each ATM node 200, and the VCI (virtual channel identifier) functions as an address for identifying the address translator 400 installed in each ATM node 200. I do.

FIG. 2 is a block diagram of the address translator 400 in the embodiment of FIG. Referring to FIG.
Data input means 1 for inputting a data packet to be used on the AN side, cell assembling means 2 for assembling the input data packet into cells, header converting means 3 for generating a header for each cell based on the input data packet; Assembly means 2
Cell generating means 4 for combining the payload of the cell generated by the above with the cell header generated by the header converting means 3 and outputting the cell as a cell to the ATM network; cell input means 5 for inputting a cell from the ATM network; Cell decomposing means 6 for decomposing the converted cells into data packets usable in the destination FLAN, address learning means 7 for associating the source data terminal address with the VPI and VCI used in the ATM network, and address learning means 7
A learning table G1 for storing the address translation information created in step (1), and a data output means 8 for outputting the data packet after completion of the decomposition to the FLAN. The data input means 1 transmits the data packet from the FLAN to the F
A function for discriminating whether data is transmitted to a destination other than the LAN and discarding the data packet when the destination is in the FLAN (filtering function), and a function of sending the data packet to the ATM network when the destination is other than the FLAN (Forwarding function). Header conversion means 3
Referring to the destination address (uniquely determined by the physical position) in the data packet from the AN, the VPI and VCI addresses of the destination corresponding to the destination address are searched from the learning table G1, and if learned, the individual It has a function of passing the cell to the cell generation unit 4 as a cell with the destination specified, and passing the cell as a cell addressed to all FLANs (hereinafter, referred to as a “broadcast cell”) to the cell generation unit 4 if the cell has not been learned. The cell input means 5 includes a header error checking function, a function of discarding a received cell when the destination of the cell is other than the own station (filtering function), and a function of discarding the received cell when the destination of the cell is the own station. A capturing function (forwarding function). The address learning means 7 determines whether the received cell is a single cell or a head cell (determination of a single cell or a head cell is “CCITT I.363-B-ISD
N Adaptation Layer (AAL)
As shown in "Specification", the source VPI and VCI in the cell header and the source address of the data packet present in the corresponding cell payload are learned in the cell header. A function for registering and updating in the table G1 is provided.

Next, a specific operation example will be described.

As shown in the cell format of FIG.
The PI field is divided into two, and the destination node address has 6 bits of VPI (DA) and the source node address has 6 bits of VPI (SA). Furthermore, V
The CI field is also divided into two, and V
CI (DA) 8 bits, VCI as source address
(SA) 8-bit configuration. An example in which the FLAN accommodated in the ATM network is a LAN of the CSMA / CD system will be described with reference to FIGS.

[0010] After the cell B1 input from the ATM network side is subjected to header check and filtering by the cell input means 5, the whole cell which has passed is passed to the cell decomposition means 6 and the address learning means 7. In the case where the received cell B1 is a single cell or a head cell, the address learning means 7 sets the VPI (SA) b3 in the cell header.
Is registered in the VPI (DA) column F2 of the learning table (G1), and the VCI (SA) b5 is stored in the VCI of the learning table G1.
(DA) column F3, and furthermore, the source address (SA) a6 (uniquely determined by the physical position) in the MAC frame in the payload of the cell b1 in the F of the learning table G1.
Register in the LAN destination address field F1. If the received cell B1 is not a single cell or a head cell, nothing is performed.
The cell disassembly means 6 assembles the payload portion of the cells in accordance with an AAL (ATM adaptation layer) type format specified in advance for the cells arriving sequentially (for the cell disassembly method, refer to "CCIT
TI. 363-B-IS DNA adaptation
The original MAC frame a4 (shown in “Layer (AAL) Specification”) is passed to the data output means 8 as soon as it is completed. The data output means 8 sends the MAC frame a4 passed from the preceding stage to the FLAN side.

After the MAC frame A1 input from the FLAN side is filtered by the data input unit 1, the entire MAC frame that has passed is passed to the cell assembling unit 2, and at the same time, the destination address (D
A) a2 is passed to the header conversion means 3. In the cell assembling means 2, a payload portion of the cell is generated in accordance with an AAL (ATM adaptation layer) type format specified in advance.
"CCITT I.363-B-ISDN Adapt"
ation Layer (AAL) Special
ation ”). In the header conversion means 3, the destination address (DA) a2 in the MAC frame a1
Is used as a search key, the FLAN destination address column F1 of the learning table G1 is referred to, an address equal to the destination address (DA) a2 is searched, and the VPI (DA) of the cell conversion information is searched.
From column F2, the destination node address VPI (DA) and from the VCI (DA) column F3, the destination address VC
Take out I (DA). Here, when the cell conversion information has already been learned, that is, an address equal to the destination address (DA) a2 in the MAC frame a1 is stored in the learning table (G
When the FPI exists in the FLAN destination address column (F1) of 1), the VPI (DA) (corresponding value in the F2 column) and VCI (DA) (corresponding value in the F3 column) of the individual destination designation are stored in the cell generation unit. Passed to 4. If the cell conversion information has not been learned, that is, if the address equal to the destination address (DA) a2 in the MAC frame a1 is F
If it does not exist in the LAN destination address field (F1), for example, VPI (DA) and VCI (DA) designated for broadcast are passed to the cell generation means 4. In the cell generating means 4, the destination addresses VPI (DA) b7 and VCI (DA) b
VPI (SA) b8 and V which are source addresses
CI (SA) b10 is added, a cell b6 is generated together with the payload from the cell assembling means 2, and the cell b6 is transmitted to the ATM network side.

FIG. 4 is a cell format diagram showing a second embodiment of the present invention in which learning is performed without dividing the VPI field into two (ie, DA and SA fields).

FIG. 5 is a block diagram exemplifying the operation when the ATM switching device 10 changes the VPI. FIG. 5 considers cell transmission from FLAN (i) accommodated in address translator 40A to FLAN (j) accommodated in address translator 40C. MA transmitted from a data terminal on FLAN (i)
The C frame is divided into a plurality of cells in the address translation device 40A. The destination node address C is set in the VPI of the divided cell header. Furthermore, VC
I (DA) is set to “3” indicating the destination address translator 40C, and VCI (SA) is set to 1 representing the source address translator 40A. This cell is AT
Received by the M-switching apparatus 10, where VPI = C indicating the destination node address is VPI indicating the source node address.
= A and sent to the destination node C. The address translator 40C that has received this cell learns the data terminal address (SA in the MAC frame) corresponding to the source VPI, VC (SA), and sends the assembled MAC frame to FLAN (j). You.

FIG. 6 is a block diagram exemplifying a case where the ATM network has a loop type configuration, and shows an operation in the case where the VPI is switched as in FIG. In FIG. 6, an ATM-D / I device 30 for taking in a cell flowing on a loop or transmitting a cell on a loop is installed instead of the ATM conversion device 10 that has performed the transfer of the VPI in FIG. Then, the transfer of the VPI is performed. Here, in order to recognize the source node address of the cell received by the ATM-D / I device 30, there is a method of recognizing the source node address from a busy token transmitted by the node that has acquired the transmission right. As described above, in the loop type ATM network, the VPI during the flow on the loop indicates the destination node, and the ATM-D / I
The VPI indicates the source node at the stage of transmission from the device 30 to the address translation device 40.

Also, as shown in the embodiment of FIG.
If a method of dividing the field into a VPI (DA) field indicating the destination node and a VPI (SA) field indicating the transmission source node is used, it is not necessary to change the VPI.

[0016]

As described above, according to the present invention, a virtual path identifier (VPI) and a virtual channel identifier (VCI) set between a source node and a destination node are considered as addresses, and the VPI, Since the VCI field is configured to be divided into a field indicating a destination address and a field indicating a source address,
Automatic learning of address translation information becomes possible, and even if a data terminal on the front-end network side is moved, added, or changed, a change operation by a system administrator is not required. Further, it is possible to easily cope with the movement or change of the data terminal that occurs daily on the front end network side.

[Brief description of the drawings]

FIG. 1 is a network configuration diagram showing one embodiment of the present invention.

FIG. 2 is a block diagram of an address translation device according to an embodiment of the present invention.

FIG. 3 is a diagram showing addresses referred to in the address conversion method of the present invention and elements of a learning table.

FIG. 4 shows VPI and VCI in a second embodiment of the present invention.
It is a cell format figure showing the example of field division.

FIG. 5 is a block diagram illustrating VPI transfer when the present invention is applied to a network having a star configuration.

FIG. 6 is a block diagram illustrating switching of VPIs when the present invention is applied to a network having a loop configuration.

[Explanation of symbols]

Reference Signs List 1 packet data input unit 2 cell disassembly unit 3 cell header conversion unit 4 cell generation unit 5 cell data input unit 6 cell assembly unit 7 address learning unit 8 packet data output unit 100 ATM switching device 200 ATM node 300 multiplexer 400 address conversion device A1 input Packet data a1 Input packet data a2 Destination address in input packet data (DA) a3 Source address in input packet data (SA) A4 Output packet data a4 Output packet data a5 Destination address in output packet data (DA) a6 Output packet data Source address (SA) B1 Input cell data b1 Input cell data b2 Destination path identifier (VPI (D
A)) b3 Source path identifier (VPI) in input cell data
(SA)) b4 Destination channel identifier (VCI) in input cell data
(DA)) b5 Source channel identifier (VC
I (SA)) B6 Output cell data b6 Output cell data b7 Destination path identifier (VPI (D
A)) b8 Source path identifier (VPI) in output cell data
(SA)) b9 Destination channel identifier (VCI) in output cell data
(DA)) b10 Source channel identifier in output cell data (V
CI (SA)) G1 Learning table F1 Destination MAC address F2 Path identifier corresponding to destination MAC address F3 Channel identifier corresponding to destination MAC address

Continuation of the front page (56) References JP-A-6-224930 (JP, A) JP-A-7-143180 (JP, A) JP-A-3-224334 (JP, A) JP-A-4-107070 (JP) JP-A-4-13331 (JP, A) JP-A-3-141754 (JP, A) JP-A-5-37569 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB Name) H04L 12/56 H04L 12/28

Claims (4)

(57) [Claims] 1. A virtual path identifier V in a header of a fixed-length packet cell comprising a header part and an information part.
By dividing the PI field and the virtual channel identifier VCI field into a destination field and a source field, respectively, the data terminal address of the source front-end network is extracted from the cell input from the ATM network side, and the header of the cell is extracted. Address learning means for storing address translation information associated with the source VPI and VCI, and retrieval means for retrieving the address translation information with reference to the destination address in the cell input from the front end network A cell generating means for generating an individual destination designation cell if already learned according to the search result and generating an unlearned cell if not yet learned, for an ATM network. Address translation method. Wherein said searching means, the search result of the address conversion information, in the case before SL cells were not learned, the cell generation means the unlearned cells as cells addressed all FLAN
The address translation method for an ATM network according to claim 1, wherein the address translation method is configured to be passed to an ATM network. 3. The VPI field indicating the source node is changed from the VPI value indicating the destination node to the VPI value indicating the source node by the device switching the cell without dividing the VPI field in the header into a destination field and a source field. 2. The address conversion method for an ATM network according to claim 1, further comprising means for performing: 4. The method according to claim 1, wherein the VPI field in the header is not divided into a destination field and a source field.
2. A according to claim 1, further comprising means for changing the PI value.
Address conversion method for TM networks.