JP2008227695A - Packet communication system and packet communication method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a packet communication system and a packet communication method capable of fully increasing the number of subscribers within a single domain. <P>SOLUTION: When a packet is transferred from a first double-VLAN tag grant apparatus 204<SB>11</SB>to a first backbone node 202<SB>1</SB>, the first double-VLAN tag grant apparatus 204<SB>11</SB>replaces a VLAN tag with a double-VLAN tag (common tag) using a tag table 232<SB>1</SB>. Prior to arriving at a destination user in-home apparatus 205<SB>205</SB>, the double-VLAN tag is replaced with the original VLAN tag at a double-VLAN tag grant apparatus 204<SB>20</SB>, using the tag table 232<SB>1</SB>. By means of the translation of tag by the tag table 232<SB>1</SB>, flexibility of tag replacement becomes large and the number of subscribers can be increased. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a packet communication system and a packet communication method, and more particularly to a packet communication system and a packet communication method suitable for transferring packets not only in individual networks but also in a wide area network connecting them.

  A VLAN (Virtual Local Area Network) can be realized by dividing each port of the switching hub into a group. With such division, VLANs can have a wide range of network configurations and can ensure security. In order to realize a VLAN function that spans between switching hubs, a special field for representing a VLAN attribute is prepared in an Ethernet (registered trademark) frame.

  FIG. 12 shows a conventional frame format when using a VLAN. This frame format is defined by IEEE (The Institute of Electrical and Electronics Engineers, Inc.) 802.3, which is a standard standard for a typical connection method of a LAN (Local Area Network). The frame 101 includes a preamble 102 from the top, a 1-byte frame start delimiter (SFD) 103 indicating the start of the frame, a destination (DA) 104, a transmission source (SA) 105, a length 106, a tag header (TAG) 107, and data 108. The frame includes a FCS (Frame Check Sequence) 109 as CRC (Cyclic Redundancy Checking) for error detection and an extension symbol CE (Carrier Extension) 110 used for a small-sized packet in Gigabit Ethernet (registered trademark).

  Here, the first two bytes of the tag header 107 are a TPID (Tag Protocol Identifier) 111. The next 2 bytes are TCI (Tag Control Information) 112, and 4-bit priority information defined in IEEE802.1P and a 12-bit VLAN-ID (Identification) for identifying the VLAN to which the frame belongs. It is set up. Since the VLAN-ID has a 12-bit configuration, a maximum of 4094 VLANs can be identified.

  The frame format shown in FIG. 12 is used in, for example, a packet communication system as a wide area Ethernet (registered trademark) service network provided by a communication carrier.

  FIG. 13 shows an example of such a conventional packet communication system. In this example, the packet communication system 120 includes a backbone network (Main Domain) 121 as a backbone network, and first and second sub-area networks (Sub Domain) 122 and 123. A first tag conversion device 124 is disposed between the backbone network 121 and the first sub-area network 122, and a second tag conversion is performed between the backbone network 121 and the second sub-area network 123. A device 125 is arranged. Also, in the first subarea network 122, as shown as representative examples, a user home device 128 is arranged via a layer 2 switch (L2SW) 126, and a user home device 129 is arranged via a layer 2 switch 127, respectively. Has been. For the second sub-area network 123, as a representative example, a user home device 133 is arranged via a layer 2 switch 131, and a user home device 134 is arranged via a layer 2 switch 132, respectively.

  In such a packet communication system 120, for example, the first sub-area network 122 can identify the VLAN-ID of each subscriber by using the frame 101 shown in FIG. However, the upper limit number of subscribers that can identify the frame 101 within a single domain is 4096 determined by the combination of 12-bit signals constituting the VLAN-ID. Therefore, the upper limit of the number of subscribers that can be accommodated in the first subarea network 122 is 4096. Therefore, in the service network shown in this figure, the number of subscribers that can be accommodated in the entire network is increased to more than 4096 by providing the second subarea network 123 in addition to the first subarea network 122. I am letting.

  It is assumed that communication is performed between the first sub-area network 122 and the second sub-area network 123 in the packet communication system 120 shown in FIG. In this case, it is necessary to transfer the frame once through the backbone network 121. At this time, a frame in which a 12-bit VLAN-ID is set is also used in the backbone network 121. Therefore, for example, if the VLAN-ID of the frame transferred from the first subarea network 122 to the backbone network 121 remains unchanged, the VLAN-ID is transferred from the second subarea network 123 within the backbone network 121. There is a possibility that the frames that have been overlapped with each other, and the two frames cannot be distinguished from each other.

  Therefore, a first tag conversion device 124 is provided for performing tag conversion of frames transferred from the first sub-area network 122 to the backbone network 121, and performs tag conversion of these frames. Similarly, when the frame transferred to the backbone network 121 is transferred to the first tag converter 124, it is necessary to convert the VLAN-ID. For this reason, the second tag converter 125 is provided. . When the frame is transferred from the second tag converter 125 to the first tag converter 124, the first tag converter 124 and the second tag converter 125 are required for the same reason.

  Since the upper limit of the number of subscribers in a single domain is limited to 4096, in order to remove this limit and increase the number of subscribers, as described above, the service network includes the first sub-area network 122. It is necessary to provide a plurality of subarea networks such as the second subarea network 123. In this case, conventionally, tag conversion devices such as the first tag conversion device 124 and the second tag conversion device 125 are arranged between the backbone network 121 and the VLAN-ID tag conversion is performed each time. There was a need to perform domain management. In other words, in a carrier with 4094 or more subscribers, it can be divided into a plurality of domains by investing enormous costs and installing a tag conversion device between each domain, and subdomains can be divided into these domains. The VLAN-ID has been expanded by an inefficient operation of transferring.

  For this reason, not only the capital investment is expensive, but the operation cost for performing complicated domain management is required, and a packet communication system that is simple and does not require the operation cost has been desired.

Therefore, at the boundary node when connecting the medium-scale network with the large-scale network, the tag value used in the large-scale network is added to the tag value used in the medium-scale network, so that the subscribers of the entire network It has been proposed to increase the number of accommodation (see, for example, Patent Document 1). In this proposal, the tag value added in the medium-scale network is set as the value of the first tag in the medium-scale network, and the value of the second tag used in the large-scale network is added to this tag. Use tags on large networks using. In addition, the value of the second tag is deleted from the packet transferred from the large scale network to the medium scale network.
Republished patent W02004 / 109987 (Claim 2, page 6, line 10 to line 30, FIG. 2)

  However, in this proposed packet communication system, the value of the first tag is provided as an independent user identifier in the medium-scale network, and the value of the second tag is used as a path identifier indicating the destination boundary node in the large-scale network. in use. For this reason, even if each tag value has a configuration of 12 bits and can have a maximum of 4096 patterns, there are not necessarily 4096 types of user identifiers and route identifiers. For example, in the extreme case, when there are only 10 types of independent user identifiers in a medium-sized network, even when there are 4096 types of route identifiers, these can be combined to identify 4096 subscribers. Only a much smaller number of VLAN-IDs can be accommodated while using a total of 24 bits.

  Therefore, an object of the present invention is to provide a packet communication system and a packet communication method capable of sufficiently increasing the number of subscribers in a single domain.

  According to the first aspect of the present invention, (a) a packet when communication is performed in each network with the first combination number as the total number of bit string combinations in the first tag having the first bit length as the maximum value. First tag allocating means for allocating a first tag for identifying each other to each packet; (b) the number of first combinations described above, and the bit string in the second tag of the second bit length Common tag assigning means for assigning a common tag for identifying packets common to all networks to each individual packet, with the maximum value obtained by multiplying the second combination number as the total number of combinations. And (c) a one-to-one correspondence between the common tag assigned by the common tag assigning means and the first tag of each network assigned by the first tag assigning means. A correspondence table creating means for creating the attached correspondence table; and (d) a first when each packet is used in one network based on the correspondence table created by the correspondence table creating means. The packet communication system is provided with tag conversion means for replacing the tag and a common tag corresponding to the first tag when used in common to all networks.

  That is, in the present invention, the first tag having the first bit length is used in each network, while the second tag having the second bit length is used as the first tag when packets are used between the networks. A common tag with the added bit structure is attached. At this time, it is necessary to associate a packet used in each network (some with a first tag and some without a tag) with a common tag common to all networks. Therefore, a correspondence table for this purpose is created, and based on this, the tag conversion means replaces the common tag when it is used in common for all networks corresponding to the first tag.

  As described above, in the present invention, since the common tag is not a simple combination of the first tag and the second tag, the subscription exceeding the number of subscribers determined by the first tag in one network (domain). The number of users can be realized using a common tag. This is a maximum of the number of subscribers determined by the first tag and the number of subscribers determined by the second tag. Therefore, the number of subscribers can be increased sufficiently within a single domain.

  Here, the correspondence table may be provided in a management server that is managed in common for all networks, or may be distributed in various places in all networks. It is also possible to create a correspondence table in which common tags are assigned to untagged packets in each network.

  In the invention according to claim 7, (a) a packet when communication is performed in each network with the first combination number as the total number of bit string combinations in the first tag having the first bit length as the maximum value. A first tag allocating step of allocating a first tag for identifying each other to each packet; (b) the number of first combinations described above; and a bit string in a second tag having a second bit length Common tag assignment step of assigning a common tag for identifying packets to each other in common to all networks, with the maximum value obtained by multiplying the second combination number as the total number of combinations. (C) the common tag assigned in the common tag assignment step and the first tag of each network assigned in the first tag assignment step. A correspondence table creation step for creating a correspondence table in which one-to-one correspondence is established, and (d) based on the correspondence table created in the correspondence table creation step, for each packet within one network The packet communication method includes a first tag when used and a tag conversion step for replacing the first tag corresponding to the first tag and used when commonly used in all networks.

  That is, according to the present invention, a first tag for identification is assigned to packets used in individual networks, and the bit length of the first tag is used for identification when these packets are used between networks. A common tag to which the bit length of the second tag is added is assigned, and a correspondence table thereof is created. When the packet is used between networks, the first tag is replaced with a common tag by referring to the correspondence table, and conversely, when the packet passing between the networks reaches the target network, it is common. The tag is replaced with the first tag. In the present invention, a tag that physically links the first tag and the second tag is not used, but a common tag associated with the correspondence table is used, so that the number of subscribers is sufficiently increased within a single domain. Can be made.

  As described above, according to the present invention, tag information is reassigned on a route through which packets are transferred using a correspondence table for tag information, so that while using packets in an existing format in individual networks, It can easily cope with the construction of a wide area network.

  Hereinafter, the present invention will be described in detail with reference to examples.

FIG. 1 shows a packet communication system as a wide area Ethernet (registered trademark) network in an embodiment of the present invention. In this packet communication system 200, the first to n-th backbone nodes 202 _{1 to} 202 _n (where n is an integer of 2 or more) configure the backbone network 201 by performing interconnection so as to form a ring. ing.

Here, the first backbone node 202 ₁ is connected to the first and second double VLAN tag assigning devices 204 ₁₁ and 204 ₁₂ for assigning and removing the double LAN tag. For example, first and second user premises devices 205 ₁₁₁ and 205 ₁₁₂ are connected to the first dual VLAN tagging device 204 ₁₁ . For example, third and fourth user home devices 205 ₁₂₃ and 205 ₁₂₄ are connected to the second dual VLAN tagging device 204 ₁₂ . The second backbone node 202 ₂ is connected to a double VLAN tagging device 204 _20, and, for example, the fifth and sixth user premises devices 205 ₂₀₅ and 205 ₂₀₆ are connected via this. Yes. The third backbone nodes _2023, double VLAN tag assigning device 204 ₃₀ is connected, an example user home apparatus 205 _307, 205 ₃₀₈ of the seventh and eighth are connected via this Yes. In the following, although not shown in the figure, the same fourth to n-th backbone nodes 202 _{4 to} 202 _n are also connected to form a service network.

Thus, in the packet communication system 200 of the present embodiment, the backbone network 201 is configured by connecting the _first to n-th backbone nodes 202 _{1 to} 202 _n in a ring shape.

In such a packet communication system 200, each user home device 205 ₁₁₁ , 205 ₁₁₂ ,... Is distinguished from the TCI 112 (hereinafter referred to as “dual VLAN tag” in the tag header 107 shown in FIG. It is assumed that a single VLAN tagged packet with a tag “)” or an untagged packet is handled. This will be specifically described the packets passing through the first double VLAN tag assigning device 204 ₁₁ for example.

The first dual VLAN tag assigning device 204 ₁₁ has a function of assigning a double VLAN tag to an untagged packet or a single VLAN tagged packet sent from the first user home device 205 _111. Yes. The first double VLAN tag assigning device 204 ₁₁ has a function of transmitting a grant double VLAN-tagged packets to the first backbone node 202 _1. Furthermore, the first double VLAN tag assigning device 204 ₁₁ reads the tag information written to the double VLAN-tagged packets received from the first backbone node 202 _1, after removal of double VLAN tag, the It has a function of transferring to the first or second user home device 205 ₁₁₁ , 205 ₁₁₂ .

First backbone node 202 ₁ reads the first double VLAN tag assigning device 204 tag information written to the double VLAN-tagged packets sent from _11, the adjacent inside the backbone network 201 according to the tag 2 has a function of transferring a packet to another backbone node 202 such as the _second backbone node 202 ₂ .

FIG. 2 shows an example of the relationship between the first backbone node and its lower devices as part of the service network shown in FIG. A packet flow will be described with reference to FIG. The first dual VLAN tagging device 204 ₁₁ is configured to send an untagged packet transmitted from the first user premises device 205 ₁₁₁ or an IEEE 802.1Q tagged packet (hereinafter referred to as an IEEE 802.1Q tag) defined for VLAN tagging. Is abbreviated as “single VLAN tag”), a 12-bit double VLAN tag is attached and transferred to the _first backbone node 2021. Further, the packet with the double VLAN tag sent from the first backbone node 202 ₁ learns the double VLAN tag and the 802.1Q tag by the first double VLAN tag adding device 204 ₁₁ . Yes.

FIG. 3 shows an outline of the first dual VLAN tagging device. The first double VLAN tag assigning device 204 ₁₁ includes a main control unit 221 for controlling the entire. The main control unit 221, a first line cards 222 ₁ provided corresponding to the first double VLAN tag assigning device 204 ₁₁ shown in FIG. 1, the second double VLAN tag assigning device 204 ₁₂ connecting the second line card 222 ₂ provided in correspondence.

The main control unit 221 includes a CPU (Central Processing Unit) 225, a control program storage medium 226 such as a ROM (Read Only Memory) storing a program to be executed, and a working memory such as a RAM (Random Access Memory). 227. The first line card 222 ₁ is connected to the main control unit 221 to add a double VLAN tag, and conversely removes the double VLAN tag processing unit 231 ₁ , and this double VLAN tag processing. A tag table 232 ₁ as a tag management database connected to the unit 231 ₁ , an up interface (I / F) circuit 233 _1, and a down interface (I / F) circuit 234 ₁ are included.

Here, the up interface circuit 233 ₁ is an interface circuit for connecting to the _first backbone node 202 ₁ shown in FIG. Also, down interface circuit 234 ₁ is an interface circuit for also connected to the first customer premises equipment 205 ₁₁₁ shown in FIG.

The second line card 222 ₂ has the same circuit configuration as the _first line card 222 ₁ . Therefore, for the second line card 222 ₂ , the subscript “1” indicating the components of the first line card 222 ₁ is used as the subscript “2”, and description of these circuits is omitted. To do.

Returning to FIG. 2, the description will be continued with FIG. The first double VLAN tag assigning device 204 ₁₁ refers to the tag table 232 ₁ and a tag table ₂₃₂₂ that reflects the learning result, performs a switching process, including double VLAN tag. Then, the double VLAN tag processing unit 231 ₁ removes the double VLAN tag field from the corresponding frame, and the down interface circuit 234 ₁ sends it to the first user home device 205 ₁₁₁ .

At this time, the packet transmitted to the first user home device 205 ₁₁₁ has been converted into an untagged packet or a single VLAN tagged packet. In this way, a packet with a double VALN tag is not sent to the first user home device 205 ₁₁₁ . Thus, the first customer premises equipment 205 ₁₁₁ without the unreachable packets sent from the first double VLAN tag assigning device 204 _11, so that it is possible to perform the receiving process.

Next, the relationship between the first backbone node 202 ₁ and the first double VLAN tag assigning device 204 _11.

FIG. 4 shows an outline of the configuration of the first backbone node. The first backbone node 202 ₁ includes a main control unit 241 that performs overall control. The main control unit 241, a first line cards 242 ₁ provided corresponding to the second backbone node 202 ₂ shown in FIG. 1, first provided corresponding to a backbone node 202 _n of the n 2 It is connected to the line card 242 _2.

The main control unit 241 includes a CPU 245, a control program storage medium 246 such as a hard disk storing a program to be executed, and a work memory 247 such as a RAM. The first line card 242 ₁ is connected to the main control unit 241 to learn a dual VLAN tag and perform a predetermined process, a dual VLAN tag learning processing unit 251 ₁ , and this dual VLAN tag learning processing unit 251 _1. a database 252 ₁ connected to is constituted by up interface circuit 253 ₁ and down interface circuit 254 _1.

Here, the up interface circuit 253 ₁ is an interface circuit for connection to the _second backbone node 202 ₂ shown in FIG. Also, down interface circuit 254 ₁ is an interface circuit for also connected to the first double VLAN tag assigning device 204 ₁₁ shown in FIG. Database 252 ₁ is adapted to perform the learning of the VLAN tag.

The second line card 242 ₂ has the same circuit configuration as the _first line card 242 ₁ . Thus, for the ₂ second line card 242, it is used as the letter "2" subscript subscript "1" in the code representing the first line card 242 ₁ component. Here up interface circuit 253 ₂ is an interface circuit for connecting to the backbone node 202 _n of the n shown in FIG. Furthermore, interface circuit 254 ₂ down, an interface circuit for also connected to the second double VLAN tag assigning device 204 ₁₂ shown in FIG. Note that the same contents may be written in the database 252 ₁ and the database 252 ₂ as learning effects.

Returning to FIG. 2, the description will be continued with FIG. When the first backbone node 202 ₁ receives the packet with the double VLAN tag sent from the first double VLAN tag assigning device 204 ₁₁ , the double VLAN tag learning processing unit 251 ₁ sets the double VALN tag. I learned to be reflected in the database 252 _1. Then, switching processing between the backbone nodes 202 is performed based on the database 252 ₁ . For example, the sent double VLAN tagged packet is forwarded to the adjacent second backbone node 202 ₂ shown in FIG. 1 using the up interface circuit 253 ₁ .

In addition, when the first backbone node 202 ₁ receives a packet with a double tag transferred from, for example, the second backbone node 202 ₂ as another backbone node, the first backbone node 202 ₁ learns a double VLAN tag. Then, switching processing is performed based on the contents of the database 252 ₁ , and the packet is transferred to, for example, the first dual VLAN tagging device 204 ₁₁ .

  As described above, double VLAN tag assignment and learning are repeated in the entire service network. As a result, a switching process using a 24-bit dual VLAN capable of identifying up to 4094 × 4094 subscribers is possible within a single domain, using a 12-bit VLAN-ID that identifies up to 4094 VLANs. become.

5 to 7 show frame formats of various packets used in this packet communication system. Among these, FIG. 5 shows a frame format 261 of a packet without a VLAN tag. FIG. 6 shows a frame format 262 for a packet with a single VLAN tag. These two types of packets are, for example, user premises equipment 205 such as first and second user premises equipment 205 ₁₁₁ and 205 ₁₁₂ in FIG. 1 and first dual VLAN tagging equipment 204 ₁₁ in FIG. The dual VLAN tagging device 204 is handled.

On the other hand, the frame format 263 for the double VLAN tagged packet shown in FIG. 7 is uniquely used in the present invention. The frame format 263, for example, to be treated in the first double VLAN tag assigning device 204 ₁₁ or device other than the user home apparatus 205 as backbone nodes 202 ₁ to 202 _n of the first to n in FIG. 1 It has become.

  Among the above three types of packets, the frame format 261 of the VLAN-untagged packet shown in FIG. 5 is a frame format of a packet that flows on a normal Ethernet (registered trademark) frame. This frame format 261 includes a destination MAC (Media Access Control) address field (DMAC) 271, a source MAC address field (SMAC) 272, an Ethernet (registered trademark) type field (Type) 273, a payload (Payload) 274, and a frame. It consists of a check sequence field (FCS) 275. Here, the transmission destination MAC address field 271 has a 6-byte configuration, and the transmission source MAC address field 272 also has a 6-byte configuration. The Ethernet (registered trademark) type field 273 has a 2-byte structure. The payload 274 has a length of 46 bytes or more, and the frame check sequence field 275 has a 4-byte structure.

  The frame format 262 for the single VLAN tagged packet shown in FIG. 6 is between the source MAC address field 272 and the Ethernet (registered trademark) type field 273 in the frame format 261 of the VLAN non-tagged packet shown in FIG. A 4-byte VLAN tag 281 is added. 12 bits in the VLAN tag 281 are assigned to the VLAN-ID.

  On the other hand, the frame format 263 for the packet with the double VLAN tag shown in FIG. 7 includes the source MAC address field 272 and the Ethernet (registered trademark) type field 273 in the frame format 261 of the packet without the VLAN tag shown in FIG. A single VLAN tag 291 and a double VLAN tag 292 having the same configuration as the VLAN tag 281 are arranged between them. Both the single VLAN tag 291 and the double VLAN tag 292 have a 4-byte configuration, and each has a 12-bit long VLAN-ID.

In this embodiment, the first double VLAN tag assigning device 204 of ₁₁ such double VLAN tag assigning device 204 and a first backbone node 202 ₁ and the like of the backbone nodes 202 shown in FIG. 2, a double VLAN tag 292 Can be handled simultaneously with the field of the single VLAN tag 291. The backbone node 202 and the dual VLAN tagging device 204 can simultaneously execute learning regarding both fields and management of the VLAN-ID.

Such learning and management, taking first double VLAN tag assigning device 204 ₁₁ shown in FIG. 3 as an example, the conventional devices are connected nonexistent double VLAN tag processing unit 231 ₁ to This is realized by the tag table 232 ₁ , the dual VLAN tag processing unit 231 ₂ and the tag table 232 ₂ connected thereto. Further, taking the first backbone node 202 ₁ shown in FIG. 4 as an example, a dual VLAN tag learning processing unit 251 ₁ , a database 252 ₁ connected thereto, and a dual VLAN tag learning processing unit 251 ₂ Such learning and management is realized by the database 252 ₂ connected thereto.

FIG. 8 shows a state in which a packet is transferred from a user home device to another user home device belonging to a network different from the network to which the user home device belongs. Here, a case will be described as an example where a packet is sent from first user home device 205 ₁₁₁ shown in FIG. 1 to fifth user home device 205 ₂₀₅ .

In this example, it is assumed that the value “10” is assigned to the first user home device 205 ₁₁₁ as the VLAN tag 281 (FIG. 6) in order to distinguish the VLAN. Packet attached with this VLAN tag 281 is input to the first double VLAN tag assigning device 204 _11, through the first down interface circuit 234 ₁ of the line card 222 ₁ shown in FIG. 3 double VLAN tag It is input to the processing unit 231 _1. Then, it sent a double VLAN tag processing unit 231 _1, wherein the frame format 262 for single VLAN-tagged packet shown in FIG. 6 to the frame format 263 for double VLAN-tagged packet shown in FIG. 7 Format conversion is performed. At this time, the dual VLAN tag processing unit 231 ₁ refers to the tag table 232 ₁ and searches for the value “10” of the VLAN tag 281 (FIG. 6), and the single VLAN tag 291 and the double VLAN tag 292 (FIG. 6). The value of 7) is changed to the value “100-12” of the common tag assigned with these total bit lengths.

The packet whose format is converted to the frame format 263 for the packet with the double VLAN tag is sent to the first backbone node 202 ₁ through the up interface circuit 233 ₁ . In the first backbone node 202 ₁ , the dual VLAN tag learning processing unit 251 ₁ shown in FIG. 4 refers to the database 252 ₁ to determine the backbone node to which this packet is sent. Then, the packet of the value “100-12” of the dual VLAN tag (common tag) is sent to the target second backbone node 202 ₂ via the backbone network 201.

In the second backbone node 202 ₂ , the dual VLAN tag learning processing unit 251 ₁ similar to the _first backbone node 202 ₁ shown in FIG. 4 refers to the database 252 ₁ . Then, the packet as it is, and sends toward the double VLAN under a frame format 263 for tagged packets double VLAN tag assigning device 204 _20.

Double VLAN tag assigning device 204 ₂₀ has the same configuration as the first double VLAN tag assigning device 204 ₁₁ of FIG. The dual VLAN tagging device 204 ₂₀ sends this packet received from the second backbone node 202 ₂ to the dual VLAN tag processing unit 231 ₁ , and refers to the tag table 232 ₁ . Then, the value “100-12” of the double VLAN tag is converted into the value “10” of the VLAN tag 281 (FIG. 6), and the frame format 262 for the packet with the single VLAN tag shown in FIG. Convert. At this time, the 24 bits assigned to the dual VLAN tag (the number of bits of the single VLAN tag 291 and the dual VLAN tag 292 (FIG. 7)) are reduced to 12 bits assigned to the VLAN tag 281 (FIG. 6). The reconverted packet is sent to the fifth user home device 205 ₂₀₅ .

In this way, it is possible to send packets by VLAN. In addition, the first to n-th backbone nodes 202 _{1 to} 202 _n use the frame format 263 for the double VLAN tagged packet shown in FIG. Therefore, the transfer destination of the frame does not become unknown.

FIG. 9 shows the flow of the packet transfer process described in FIG. 8 in the packet communication system of the present embodiment. This will be described with reference to FIG. When single VLAN-tagged packets from the first customer premises equipment 205 ₁₁₁ is sent (step S301), the first double VLAN tag assigning device 204 _11, by searching the tag table 232 ₁ (step S302), the results imparting double VLAN tag in double VLAN tag processing unit 231 ₁ using (step S303). At this time, if there is no double VLAN tag corresponding to the tag table 232 ₁ , the packet is discarded as an error. When a double VLAN tag is added, the configuration and contents of the packet bit string change accordingly. Therefore, the frame check sequence (FCS) is rewritten by recalculating CRC (Cyclic Redundancy Checking) from the top of the frame.

The packet with the dual VLAN tag is transferred to the _first backbone node 2021 of the backbone node, and the dual VLAN tag is read (step S304). Then, retrieved by the database 252 ₁ (step S305), and transferred to the destination node, such as a neighbor node if other backbone nodes are destined (step S306).

In this example, since the second backbone node 202 ₂ is the target node, reading of the double VLAN tag (step S307) and searching of the database 252 ₁ are performed here (step S308). Then, when there is a match to be transferred to the corresponding double VLAN tag assigning device 204 ₂₀ (step S309). If there is no match in the search in step S305 and step S308, the packet is discarded.

In the dual VLAN tag assignment device 204 ₂₀ , a search using the tag table 232 ₁ is performed (step S 310), and the dual VLAN tag processing unit 231 ₁ removes the dual VLAN tag. At this time, the FCS is rewritten by recalculating the CRC from the beginning of the frame. In this way, the packet has the frame format 262 for the single VLAN tagged packet shown in FIG. 6 (step S311), and the fifth user home device 205 ₂₀₅ receives the packet in this format (step S312). ).

Finally, a method of creating a tag conversion table that is the basis of the tag table 232 and the database 252 in this embodiment will be briefly described. Individual This Field of the common tag and, like the first customer premises equipment 205 ₁₁₁ to use a region of a single use double VLAN tag 291 and double VLAN tag 292 (FIG. 7) in the backbone network 201 shown in FIG. 1 This is a method for creating a conversion table for conversion between VLAN tags 281 (FIG. 6) used in the network.

FIG. 10 is a generalized representation of the creation of this conversion table. First, the first combination number as the total number of combinations of bit strings in the first tag having the first bit length is set as a maximum value, and a first for identifying packets when communicating in each network. Are assigned to individual packets scheduled to be used (step S331). In the present embodiment, the first tag in the first tag assignment step is the VLAN tag 281 in the frame format 261 of the VLAN tag-less packet shown in FIG. 5 or the double VLAN tag attached as shown in FIG. This means a single VLAN tag 291 in the frame format 263 for the packet. In this embodiment, the number of bits for the combination is 12 bits, and the first combination number is 4096. The maximum of 4096 first tags are allocated to individual packets scheduled to be used in individual networks to which the first user home device 205 _{111 and the} like belong. This is the same as in the prior art.

Next, the maximum number is obtained by multiplying the maximum number of 4096 first combinations by the second number of combinations as the total number of bit string combinations in the second tag having the second bit length. A common tag for identifying packets common to all networks is assigned to each packet scheduled to be used (step S332). In this common tag assignment step, the second tag having the second bit length means the double VLAN tag 292 in the frame format 263 for the packet with the double VLAN tag shown in FIG. Yes. Therefore, the second combination number is 4096. For this reason, the number obtained by multiplying the number of first tag combinations and the number of second combinations in this embodiment is 4096 × 4096. The common tag means tag information using the bit areas of both the single VLAN tag 291 and the double VLAN tag 292 (FIG. 7). Therefore, in this step S332, a maximum of 4096 × 4096 common tags are assigned to individual packets scheduled to be used in individual networks to which the first user home device 205 _{111 and the} like belong.

  In the next step S333, a correspondence table in which the common tag assigned in the common tag assignment step and the first tag of each network assigned in the first tag assignment step in step S331 are associated one-to-one is created. create. In this correspondence table creation step, a correspondence table showing the correspondence relationship is created. However, the correspondence shown in the correspondence table need not be fixed permanently. At each time point, the identification information of the packet used in all networks may be associated with the identification information of the packet used only in each network.

When the correspondence table is created in this way, the dual VLAN of the first dual VLAN tag assigning device 204 ₁₁ shown in FIG. 1 is created based on the correspondence table created in the correspondence table creation step in step S333. Replacing the first tag when the tag assigning device 204 is used in one network for each packet and the common tag when used in common for all networks corresponding to the first tag Should just be done. A part of the conversion table used in this tag conversion step is, for example, a tag table 232 ₁ and a database 252 ₁ for determining a route.

In the tag table 232 ₁ and the database 252 ₁ , the correspondence table created in step S333 may be completed at once, and the related ones may be allocated in advance to each place, and what is necessary each time is required. it may be caused to store the tag table 232 ₁ and a database 252 ₁ of these places by learning. Also, unlike the case described with reference to FIG. 10, the correspondence between tag information used in one network and tag information common to all networks is shown each time a packet forwarded to an external network is generated. It is also possible to allocate and create or grow a correspondence table.

  As described above, according to the present embodiment, it is possible for the communication carrier to take in more subscribers than 4096 in a single domain. Therefore, the service can be improved as compared with the prior art. In addition, it is possible to prevent duplication of IEEE802.1Q tags that have been opened to the user's home. Thereby, the reliability as a communication carrier can be improved.

  Conventionally, in order to accommodate more than 4096 subscribers, it has been necessary to divide the domain into a plurality of domains. As a result, management within the network has become complicated, and operational costs have become enormous. In the packet communication system 200 according to the present embodiment, a maximum of 4094 × 4094 VLAN-IDs can be handled in a single domain in a wide area Ethernet (registered trademark) network. By increasing the upper limit of accommodation, it is possible to reduce the operation cost.

  Furthermore, conventionally, when dividing into a plurality of domains, a device for exchanging VLAN tags between the domains is required, which requires a large capital investment. In the present embodiment, the necessity of domain division is eliminated, so that it is possible to reduce the capital investment for the business operator and to reduce the line charge for the subscriber.

  <Modification of the invention>

  FIG. 11 shows the configuration of a packet communication system in a modification of the present invention. In FIG. 11, the same parts as those in FIG. 1 are denoted by the same reference numerals, and description thereof will be omitted as appropriate.

In the packet communication system 200A of this modified example, the dual VLAN tag management server 401 is connected to an arbitrary network of the backbone network 201. The dual VLAN tag management server 401 is connected to a dual VLAN tag management database (DB) 402 as its internal or external external recording device. The dual VLAN tag management database 402 is composed of a mass storage device such as a magnetic disk or an optical disk, and includes a first dual VLAN tag assigning device 204A ₁₁ and a second dual VLAN existing on a communication network. The dual VLAN tags of all the dual VLAN tag assignment devices 204A such as the tag assignment devices 204A ₁₂ ,... Are centrally managed.

In this modification, the regularity and setting policy regarding the provision of the double VLAN are set in advance in the double VLAN tag management server 401. Thereby, it becomes possible to easily manage the dual VLAN tag of the entire wide area Ethernet (registered trademark) network, and as a result, it is possible to reduce the operation cost. Thus, in this modification, the dual VLAN tag management database 402 as a database for VLAN management is not provided in a single device such as the first dual VLAN tag assignment device 204A ₁₁ but as a common server. In the double VLAN tag management server 401. The double VLAN tag management database 402 can be considered to be a collection of the correspondence table of the embodiment described with reference to FIG. This not only reduces the operational load on the operator, but also facilitates management.

Furthermore, in this modification, for example, when the first dual VLAN tag assigning device 204A ₁₁ is started up, the dual VLAN tag management server 401 can be automatically inquired about dual VLAN information to be managed by itself. For example, it is possible to anticipate the evolution of technology that enables more dynamic operations. In this modification, all the dual VLAN tag assignment devices 204A such as the first dual VLAN tag assignment device 204A ₁₁ , the second dual VLAN tag assignment device 204A ₁₂ ,. Therefore, it is not necessary for the dual VLAN tagging devices 204A ₁₁ , 204A ₁₂ ,... Constituting them to have a large capacity memory individually. Thereby, the effect that the cost reduction of an apparatus can be anticipated is also acquired.

1 is a system configuration diagram showing a packet communication system as a wide area Ethernet (registered trademark) network in an embodiment of the present invention. FIG. It is a principle figure showing an example of the relation between the 1st backbone node and its lower rank apparatus in a present Example. It is a block diagram which shows the outline | summary of a structure of the 1st double VLAN tag provision apparatus in a present Example. It is a block diagram which shows the outline | summary of a structure of the 1st backbone node in a present Example. It is a block diagram which shows the frame format of the packet without a VLAN tag used by a present Example. It is a block diagram which shows the frame format about the packet with a single VLAN tag used by a present Example. It is a block diagram which shows the frame format about the packet with a double VLAN tag used by a present Example. It is explanatory drawing showing a mode that a packet was transferred from the user premises apparatus which is a present Example to the other user premises apparatus which belongs to networks other than the network to which it belongs. 9 is a flowchart showing a flow of packet transfer processing described in FIG. 8. It is a flowchart which shows the preparation method of the correspondence table used in a present Example. It is a system block diagram showing the structure of the packet communication system in the modification of this invention. It is explanatory drawing showing the conventional frame format at the time of VLAN use. It is the system block diagram which showed an example of the conventional packet communication system.

Explanation of symbols

200, 200A Packet communication system 201 Backbone network 202 Backbone node 204, 204A Dual VLAN tagging device 205 User premises device 221, 241 Main control unit 225, 245 CPU
231 Duplex VLAN tag processing unit 232 Tag table 251 Duplex VLAN tag learning processing unit 252 Database 261 Frame format of packet without VLAN tag 262 Frame format of packet with single VLAN tag 263 Frame format of packet with dual VLAN tag

Claims (7)

First tag for identifying packets when communicating in each network with the first combination number as the total number of bit string combinations in the first tag having the first bit length as a maximum value First tag assigning means for assigning to each packet;
The number obtained by multiplying the first combination number and the second combination number as the total number of bit string combinations in the second tag having the second bit length is set as a maximum value, and is shared by all networks. Common tag assigning means for assigning a common tag for identifying each other to the individual packets;
Correspondence table creating means for creating a correspondence table in which the common tag assigned by the common tag assigning means and the first tag of each network assigned by the first tag assigning means are associated one-to-one;
Based on the correspondence table created by the correspondence table creating means, the first tag when used in one network for each packet, and the first tag corresponding to the first tag A packet communication system comprising: tag conversion means for replacing the common tag when used in common with a network.   2. The packet communication system according to claim 1, further comprising a management server that manages the correspondence table created by the correspondence table creation means in common to all networks.   2. The packet communication system according to claim 1, wherein the correspondence table created by the correspondence table creation unit is distributed and arranged at various locations in the entire network.   2. The packet communication system according to claim 1, wherein the correspondence table creating means creates a correspondence table in which the common tag is assigned to untagged packets in the respective networks.   The packet communication system according to claim 1, wherein the tag for identifying the packets is a VLAN tag.   2. The packet communication system according to claim 1, wherein each of the first bit length and the second bit length is 12 bits, and each network handles a VLAN-ID of 4094 × 4094 at the maximum. . First tag for identifying packets when communicating in each network with the first combination number as the total number of bit string combinations in the first tag having the first bit length as a maximum value A first tag assignment step for assigning to individual packets;
The number obtained by multiplying the first combination number and the second combination number as the total number of bit string combinations in the second tag having the second bit length is set as a maximum value, and is shared by all networks. A common tag assignment step for assigning a common tag for identifying each other to the individual packets;
A correspondence table creation step for creating a correspondence table in which the common tag assigned in the common tag assignment step and the first tag of each network assigned in the first tag assignment step are associated one-to-one;
Based on the correspondence table created in the correspondence table creation step, the first tag when used within one network for each packet, and the first tag corresponding to the first tag A packet communication method comprising: a tag conversion step of replacing the common tag when used in common with a network.

Images