JPH06261078A - Table retrieval method and router - Google Patents

Abstract

PURPOSE:To quicken route processing by a router. CONSTITUTION:Each of plural network interfaces 110 in a router 100 has a filtering control section 120 and a routing control section 130, the filtering control section 120 has a filtering table 140 and the routing control section 130 has a CAM 151 having a retrieval key, a cross reference table 152 having a painter and a routing table 154. A destination address requiring lots of retrieval chances is registered in a cache memory by providing the CAM 151 and the cross reference table 152 for the routing table retrieval cache memory 150. Thus, the table retrieval is quickened and the route processing capability is enhanced. Moreover, the limitation to an entry size or the like by the CAM is solved through the provision of the cross reference table.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a routing table search system in a router of a communication system.

[0002]

2. Description of the Related Art A plurality of LANs and WANs to which a plurality of terminals are connected, and a router having a bridge function having a plurality of network interfaces for interconnecting these terminals (sometimes referred to as bruta or bridging router) A router that receives communication information between terminals connected to different networks must relay (route) the information to other networks. Therefore, it has a database in which the destination network address, the network address of the next router for reaching the network having the network address, and the number of the network interface to be sent out are one entry.

In recent years, the transmission speed of the network has become higher and higher, and the larger scale of the network has required higher speed of the routing process. As a method for speeding up these processes, for example, a paper "16 kbit string search engine LSI for processing LAN address filtering at an extremely high speed" (Journal of the Institute of Electronics, Information and Communication Engineers, May 1989, Vol. J72-C-II No.
There is a method described in 5). The above-mentioned prior art shows a high-speed search method for a filtering table in a bridge, in which the filtering table itself is registered in an associative memory (hereinafter, referred to as CAM: Content Addressable Memory), and the filtering table registration / search process is performed by a hard disk. This is done by software to speed up the filtering process.

[0004]

In the case of the above prior art,
Although the filtering process can be speeded up, since the filtering table itself is registered in the CAM, the number of entries that can be registered is limited to the number of entries in the CAM. Even when it is applied to the routing table of the router, the number of entries in the routing table is increasing due to the increase in the scale of the network, so that it is necessary to increase the capacity of the CAM. The number of entries in the CAM is relatively small at the present time, 256 and 512, and increasing the capacity is an issue for the future. Also, depending on the routing protocol, the key information for searching the routing table may be
It may be larger than the CAM entry size.

An object of the present invention is to take advantage of the speed-up of search processing, which is an advantage when using CAM, and to supplement the limitations on the number of entries and entry size, which are currently disadvantages, to make a routing table, a filtering table, or an address. It is to provide a method for accelerating the resolution table search processing.

[0006]

In order to achieve the above object, the present invention uses a CAM as a part of a cache memory in a router device as a routing control unit, a filtering control unit or an address resolution control unit. Characterize. For example, in the routing control unit, the destination address is extracted from the received packet, the routing table is searched once using the destination address as a key,
A destination address whose pointer to the corresponding entry on the routing table is known is registered in the CAM. Then, the pointer to the corresponding entry of the routing table is registered in the correspondence table having the entry corresponding to the entry of the CAM one to one.

[0007]

According to the present invention, when the table is searched again for the same destination address, if the destination address is collated with the CAM, it is instantly known whether or not the address is registered. By obtaining the pointer to the table from the entry of the correspondence table corresponding to the entry in which the corresponding address of the CAM is registered, it is possible to speed up the table search.

As described above, by providing the CAM and the correspondence table as the cache memory for searching the routing table, the filtering table, or the address resolution table, the destination address that is frequently searched is registered in the cache memory. The search speed can be increased and the routing processing capacity can be improved. Further, by providing the correspondence table, it is possible to solve the constraint such as the entry size by the CAM.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described below with reference to FIGS.
Will be explained. The first embodiment is characterized by the search of the routing table in the router. First, in Figure 1,
The structural example of the router to which this invention is applied is shown. In the figure, reference numeral 100 is a router, and the network interfaces 110 (110A to 110D) of the routers are filtering control units 120 (120A to 120D).
D), the routing control unit 130 (130A to 13)
It is divided into 0D). Filtering control unit 120
Are the filtering tables 140 (140A
Through 140D, except for 140A in FIG. 1, which is the same hereinafter). In addition, the routing control unit 130 has the CAM 151 and the correspondence table 1 respectively.
It has a cache memory 150 composed of 52, a routing table 154, and a packet transmission / reception buffer 156. 160 is a router management unit, 162 is a spare unit, 17
Reference numeral 0 indicates a management device.

Terminals 191 and 192 connect to a local area network (LAN) 181, and terminals 193 and 194 connect to L.
LAN181 and 18 connected to AN182 respectively
2. Wide area network (WAN) 183 and 18
4 is a network interface 1 in the router 100
10A to 110D, respectively.

FIG. 2 shows a network interface 110.
The structural example of A is shown. The filtering control unit 120A
The network transmission / reception circuit 210A for transmitting / receiving packets to / from the LAN 181 and the reception buffer 2 of the routing control unit 130A by searching the filtering table 140A.
It has a filtering control circuit 220A that determines whether to pass it to 30A or discard it. Routing control unit 1
The 30A adds necessary information to the packet in order to search the routing table 154A through the CAM 151A and the correspondence table 152A for the packet stored in the reception buffer 230A and to send it to another network interface (for example, 110B). It has a routing control circuit 240A for determining whether to pass it to the inter-transmission / reception circuit 250A or discard it, and a transmission buffer 260A for transmitting to the LAN 181 via the network transmission / reception circuit 210A. These circuits are controlled by the processor 270A according to a program in the memory 280A.

The router 100 shown in FIGS. 1 and 2 is a network layer which is a third layer of an OSI (Open Systems Interconnection) reference model of ISO (International Organization for Standard). However, when the router 100 receives the information of the protocol that is not supported by the third layer, the routing cannot be performed. Therefore, the routing is performed by the data link layer which is the second layer of the OSI reference model. (Bridge processing). FIG. 3 shows a branch of the router process and the bridge process of the router 100. When the router 100 receives the packet exchanged between the terminals, it extracts the protocol type indicating the protocol of the network layer from the packet (step 310), and if the extraction is possible (320), the extracted protocol type is supported by the router 100. It is judged whether the protocol is present (330), if it is supported, router processing is performed (340), and if it is not supported and if the protocol type cannot be extracted in step 320, bridge processing is performed (350). . The present invention is effective for both router processing and bridge processing,
In the following, first, the router process 340 will be focused and described.

FIG. 4 shows an example of the format of the routing table 154. It has a plurality of entries 410 for storing a destination network address, a mask for the destination network address, a destination network address in the sending network, and a sending interface number.

FIG. 5 shows an example of the format of the filtering table 140. An entry 510 that stores a data link layer address (hereinafter referred to as a physical address)
Have multiple.

FIG. 6 shows a cache memory 150 (CAM1
51 and a correspondence table 152). The CAM 151 has a plurality of entries 610 which store destination network addresses. Correspondence table 152
Is the destination network address, routing table 1
It has a plurality of entries 620 each of which stores a pointer indicating which entry 410 in 54 corresponds to. CAM
Each entry of 151 (for example, 610) has a one-to-one correspondence with each entry (for example, 620) of the correspondence table 152.

FIG. 7 shows an example in which a router to which the present invention is applied is connected to a network. The terminals 731 and 732 are connected to the network 721, and the network 722
A terminal 733 is connected to the network 721 and 72.
2 is a network interface 110 of the router 100
A and 110B are interconnected. A terminal 734 is connected to the network 723, and the networks 722 and 723 are interconnected via the network interfaces 710A and 710B of the router 700 similar to the router 100. For these network systems, network addresses that are unique within the system are defined, and the networks 721 to 723 and the router 10 are defined.
0 and 700 network interfaces 110A,
110B, 710A, 710B, terminals 731-73
Assign to 4. This network address is OSI
This is the address handled by the network layer of the reference model. A network address consists of 32 bits, for example, and it is general to divide it into four groups of 8 bits and express it in decimal number. The network addresses 741 to 743 shown in FIG. 7 follow this notation.

The network address is a field indicating the network number to which it belongs and each element in the network (networks 721 to 723, router 1).
00 and 700 network interface 110
A, 110B, 710A, 710B, terminals 731-734). In the example of FIG. 7, the upper 16 bits are the network number and the lower 16 bits are the element identifier. The network address 741 of the network 721 is 133.144.0.0, which means that the network number is 133.144 and the element identifier is 0.0.

Network interfaces 110A, 110B, 710A, 710 of routers 100 and 700
B, each of the terminals 731 to 734 is provided with a physical address handled by the data link layer of the OSI reference model, in addition to the above network address.
This physical address is information valid only within the network to which it belongs and is used for routing control inside the network.

FIG. 8 shows network addresses and physical addresses assigned to the network system shown in FIG. The physical address is 16 bits or 48 bits, but has an appropriate width in order to briefly explain the operation of the embodiment.

With the network system configuration shown in FIG. 7, by allocating the addresses shown in FIG. 8, FIG. 9 shows an example of the routing table 154A and the filtering table 140A of the network interface 110A of the router 100. Other network interfaces also have tables similar to 154A and 140A. The contents of the routing table 154A are
It may be dynamically generated by a routing protocol executed between the routers 100 and 700, or may be statically generated by being registered in advance by an administrator.

In the routing table 154A, which network interface of a particular network can be reached, and when a packet is transmitted through the network interface, the destination of the destination in the transmission network is determined. Information is written. Destination in sending network = “Nu
"ll" indicates that the local router is connected to the corresponding network.

Next, the operation of the first embodiment of the present invention will be described with reference to FIGS. 9 to 14, focusing on the network interfaces 110A and 110B. FIG. 9 shows the contents of the routing table 154A and the filtering table 140A in the network interface 110A before the operation is started.

FIG. 10 shows the network interface 11
An example of the packet received by 0A is shown. (A) is a packet 101 for communication from the terminal 731 to the terminal 733.
0 and (b) show a packet 1020 for communication from the terminal 732 to the terminal 734. TYPE is a protocol type, and FCS is a frame check sequence.

FIG. 11 shows the CA of the cache memory 150A.
The contents of the M151A and the correspondence table 152A and the contents of the address resolution table 1150 for mapping the network address and the physical address of the network interface 110B are shown. 12 is a processing flow of the filtering processing unit 120A, and FIG. 13 is a routing processing unit 13 until the end of the routing table search.
The processing flow of 0A is shown. FIG. 14 shows an example of a packet transmitted from each of the network interfaces 110B and 710B.

In the following description, the operation of the network interface 110A from receiving the packet shown in FIG. 10 to reaching the destination terminal will be described. (1) When receiving the packet 1010 As shown in FIG. 12, the filtering control unit 120A determines the destination physical address DA (=
0001) is extracted (step 1210), the filtering table 140A of FIG. 9 is searched using the DA as a key (1220), and since there is no entry whose physical address matches DA (1230), the packet 1010 is stored in the reception buffer 230A. It is stored (1250).

As shown in FIG. 13, the routing control unit 130A determines the destination network address DN (= 133.14) from the packet 1010 stored in the reception buffer 230A.
4.0.3) is extracted (1310), and DN is used as a key in Fig. 1.
A search request is sent to the CAM 151A of No. 1 by hardware (1320). When there is an entry in which the DN and the destination network address match (1330), it is checked whether the destination network address registered in the correspondence table matches the DN (1340). If they match, a pointer RP to the routing table 154A
Is acquired (1350).

Next, the routing table 1 pointed to by the RP
Value MN (= 133.145.0.) Obtained by multiplying DN (= 133.145.0.3) by the mask (= 255.255.0.0) of entry 910 of 54A.
0) is calculated (1360). Since the MN matches the destination network address (= 133.145.0.0) in the entry 910 (1370), this entry 910 is used as the search result in the routing process.

If there is no entry in which the DN and the destination network address match in step 1330 or the destination network address registered in the correspondence table in step 1340 does not match the DN, the MN in step 1370. , Does not match the destination network address in the entry 910, a cache miss hit occurs, and the routing table 154A is directly searched using the DN as a key (1380).

The packet 1010 is transferred to the transmission interface 110B indicated by the entry 910 according to the search result of the routing control unit. Since the destination in the sending network is Null, the terminal 733
Exists on the network connected to the network interface 110B, the address resolution table 1150 (FIG. 11) is directly searched by the destination network address (= 133.145.0.3) of the terminal 733 and the destination physical address (= 000C) is searched. To get Therefore, the packet 1410 shown in FIG. 14A is transmitted from the network interface 110B, and the packet 1410 reaches the destination terminal 733.

(2) When receiving the packet 1020 The same routing table search as in (1) is performed, and the entry 920 shown in FIG. 9 is used as the search result in the routing process. This result indicates that the packet 1020
Is the outgoing interface 110 shown in entry 910
B. The destination in the sending network is 133.14
Since it is 5.0.2, the terminal 734 does not exist on the network connected to the network interface 110B. Therefore, the address resolution table 1 of FIG.
150 is the destination in the sending network (= 133.145.0.
The destination physical address (= 0003) is obtained by searching in 2). Therefore, the network interface 110B
Transmits a packet 1420 shown in FIG. 14B.

This packet 1420 is received by the network interface 710A, processed in the same manner as the above-mentioned network interface 110A, transferred to the network interface 710B, and the packet 1430 shown in FIG. 14C from the network interface 710B. This packet 1430 is transmitted and reaches the destination terminal 734.

Next, the network interface 110
Pay attention to A, and use FIG. 6 and FIG. 9 to FIG.
The registration method of M and the correspondence table will be described. The CAM and the correspondence table before receiving the packet 1010 shown in FIG. 10 are in a state in which nothing is registered in the format as shown in FIG. Routing table 154
It is assumed that A and the filtering table 140A have the contents shown in FIG. FIG. 15 shows a processing flow of the routing processing unit 130A.

In the following description, the operation when the network interface 110A receives the packet 1010 shown in FIG. 10 will be described. As shown in FIG. 12, the same operation as that of the first embodiment is performed until the packet 1010 is stored in the reception buffer 230A. FIG.
As shown in, the cache mishit processing of step 1380 is performed when nothing is registered in the CAM and the correspondence table.

As shown in FIG. 15, the routing controller 130A extracts the destination network address DN extracted from the packet 1010 stored in the reception buffer 230A.
The routing table 154A is searched using (= 133.144.0.3) as a key (step 1510), the contents are read for each entry, and the search is performed until an entry whose masked DN information matches the destination network address is found. As a result, if there is a matching entry (1520), the pointer RP of this entry is acquired (1530), and it is checked whether or not there is a free entry in the CAM 151. If there is a free entry (1540), the DN is registered in that free entry. Then (1550), the DN and RP are registered in the entry of the correspondence table corresponding to that entry (1560). In step 1520, if no matching entry is found, registration is not possible. If there is no free entry in the CAM in step 1540, one of the in-use entries is made free by using an appropriate algorithm (1570), and the DN is registered in that entry (1550).

Next, a second embodiment of the present invention will be described with reference to FIG. The second embodiment is C in the first embodiment.
It is a modification of the registration method of the AM and the correspondence table. That is, regarding the operation at the time of registering the CAM and the corresponding table and at the time of searching the routing table when the destination network address which is the search key for searching the routing table 154 is larger than the entry size of the CAM 151, the routing control unit of FIG. This will be described using the processing flow of. This process is performed by step 153 in FIG.
This is a processing flow in which processing after 0 is replaced.

As shown in FIG. 16, a CAM free entry is found (steps 1540 and 1570), and the hash function H for converting the destination network address DN into the CAM entry size is used to hash the DN to the destination network address HN. (1610),
This HN is registered in the empty entry of the CAM (162
0), the DN and the pointer RP to the routing table are registered in the corresponding entry of the correspondence table (156).
0).

A third embodiment will be described with reference to FIG. In this example, steps 1320 to 13 of FIG.
It replaces the processing of 30. That is, in FIG.
In step 1310, the destination network address DN extracted from the packet is converted into an HN by a hash function (1710), and the HN is used as a key to issue a hardware search request to the CAM (1720). It is checked whether there is a matching entry (1730), and if there is an entry, the processing from step 1340 of FIG. 13 is performed, and if there is no entry, the cache mishit processing of step 1380 of FIG. 13 is performed. This cache mishit processing is the processing flow shown in FIGS. 15 and 16.

Next, a fourth embodiment of the present invention will be described with reference to FIG. This embodiment is a modification of the registration method of the CAM 150 and the correspondence table 152. As described in the first embodiment, the destination network address is used as the search key for searching the routing table 154. However, depending on the type of routing protocol,
The destination network address and other information may be used. This other information is, for example, a communication quality service (TOS: Type Of Service, or QOS).
OS: called Quality of Service). Specifically, when there are a plurality of routes for the destination, the above-mentioned routing protocol that enables setting of the communication quality service or the like is applicable. In this case, the routing table has a plurality of entries having the same destination network address but different communication quality service values. FIG. 18 shows the routing table 15 when the destination network address and TOS are used as search keys.
4 and CAM 151 and an example of the format of the correspondence table 152, each of a plurality of entries 1810 to 183.
Has 0.

When there are a plurality of search keys, they are combined as a character string into one key, which is replaced by the destination network address which is the search key described in the first to third embodiments. The operation is clear. Even if the combined key is larger than the CAM entry size, it can be solved by the third embodiment.

Next, a fifth embodiment of the present invention will be described with reference to FIGS.
This will be described using 1. The fifth embodiment shows a search method and a registration method of the address resolution table. In the description of the operations of the first and second embodiments, the use example of the address resolution table is shown in the flow of a series of operations, but in the fifth embodiment, the search and registration of the address resolution table is performed. The operation of the CAM and the correspondence table will be described.

In FIG. 19, only the routing control unit of the network interface configuration diagram shown in FIG. 2 is extracted and an address resolution table and its control circuit are added. The address resolution control circuit 1910A includes an address resolution table 1924A that is a mapping table of network addresses and physical addresses, and a CAM that stores a cache memory 1920A, that is, a search key.
It controls the correspondence table 1922A that stores pointers to 1920A and the address resolution table.

The operation after the routing table search in the operations of the first and second embodiments will be described in the fifth embodiment. 20A shows the contents of the address resolution table (the same as 110B in FIG. 11), and FIG. 20B shows the contents of the CAM and the correspondence table. Figure 21
Is the routing processing unit 1 after searching the routing table
The processing flow of 30A is shown.

When the packet 1010 is received, FIG.
Processing step 1370 and processing step 1 of FIG.
After 560, using the destination network address DN1 (= 133.145.0.3) of the terminal 733 as a key, the CA in FIG.
A hardware search request is issued to M1921A (step 2105). When there is an entry (2040) in which the destination network address matches DN1 (211)
0), it is checked whether the destination network address (= 133.145.0.3) registered in the correspondence table matches DN1 (2115). Since they match, pointer RP1 (= 1002) to address resolution table 1924A.
Is acquired (2120). Address resolution table 192
The destination physical address (= 000C) is obtained from the entry 2020 of the corresponding pointer of 4A. On the other hand, processing step 2
If NO at 110 or 2115, a cache miss hit occurs, and the same process as the process at the time of the cache miss hit shown in FIG. 15 is performed (2125 to 2125).
155).

Next, a sixth embodiment of the present invention will be described with reference to FIGS.
4 will be described. The sixth embodiment shows a search method and a registration method of the filtering table of the routing control unit when the router device also has a bridge function. The embodiment shown in FIG. 22 is obtained by adding a filtering table and its control circuit to the configuration of the routing controller of FIG. Bridge control circuit 2210A
Is a filtering table 2224A which is a mapping table of physical addresses and transmission interfaces,
Cache memory 2220A (CA storing search key
M2220A, a correspondence table 2222A) that stores a pointer to the filtering table. FIG. 23
23A shows the contents of the filtering table, and FIG. 23B shows the contents of the CAM and the correspondence table 2220A. FIG. 23C is a configuration example of the network system of FIG. 7, and shows an example of a packet subjected to bridge processing from the terminal 731 to the terminal 733. FIG. 24 shows a processing flow of the bridge processing (processing step 350 in FIG. 3) in the routing processing unit 130A. However, in FIG. 24, the description of the learning process of the filtering table 2224A itself is omitted.

When the packet 2370 is received, the packet 23 is divided in the branch of the router process and the bridge process of FIG.
Assuming that the 70 TYPE field is not supported, the bridging process (step 350) is performed. Destination physical address DA of terminal 733 (= 000C)
22 is used as a key to issue a search request to the CAM 2221A of FIG. 22 by hardware (step 2405), and there is an entry (2340) whose DA and the destination physical address match (2410), and the destination physical registered in the correspondence table. It is checked whether the address (= 000C) matches DA (2415). Since they match, pointer RP3 to filtering table 2224A (= 2002).
Is acquired (2420). Filtering table 22
The transmission interface (= 110B) is obtained from the entry 2320 of the corresponding pointer of 24A. Processing step 24
If NO in 10 or 2415, a cache miss hit occurs and the same process as the process at the time of the cache miss hit shown in FIG. 15 is performed (2425 to 2425).
55). In the case of the bridge processing, if there is no entry having a matching DA and destination physical address (2430), the sending interface is all interfaces (broadcast), and if there is an entry having a matching DA and destination physical address (2430). Then, if the sending interface is equal to the received interface (= 110 A), the packet is discarded.

[0046]

According to the present invention, by providing a CAM and a correspondence table as a cache memory for searching a routing table, a filtering table or an address resolution table, a destination address with a large number of searches is registered in the cache memory. . As a result, it is possible to speed up the table search and improve the capability of routing processing and the like. Further, restrictions on the entry size and the like due to the CAM can be solved by providing the correspondence table.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a router according to a first embodiment of the present invention.

FIG. 2 is a diagram showing details of a network interface configuration of FIG.

FIG. 3 is a diagram showing a branch of a router process and a bridge process in the apparatus of FIG.

FIG. 4 is a diagram showing an example of a routing table of FIG.

5 is a diagram showing an example of a filtering table of FIG.

FIG. 6 is a diagram showing an example of formats of a CAM and a correspondence table of FIG. 1.

FIG. 7 is a diagram showing a configuration example of a network system of a first embodiment.

8 is a diagram showing address allocation of the network system of FIG.

FIG. 9 is a network interface 1 of the first embodiment.
It is a figure which shows an example of the content of the table which 10A has.

FIG. 10 is a diagram showing an example of a packet received by the network interface 110A of the first embodiment.

FIG. 11 is a diagram showing an example of contents of a CAM, a correspondence table, and an address resolution table according to the first embodiment.

FIG. 12 is a diagram illustrating a processing flow of a filtering control unit according to the first embodiment.

FIG. 13 is a diagram showing a processing flow of a routing control unit of the first embodiment.

FIG. 14 is a diagram showing an example of packets transmitted from the network interfaces 110B and 710B of the first embodiment.

FIG. 15 is a diagram showing a registration processing flow of a CAM and a correspondence table according to the first embodiment.

FIG. 16 is a diagram showing a processing flow of a routing control unit of the second embodiment.

FIG. 17 is a diagram showing a processing flow of a routing control unit of the third embodiment.

FIG. 18 is a diagram showing an example of the contents of a table of the fourth embodiment.

FIG. 19 is a diagram showing a configuration diagram of a routing control unit of the fifth embodiment.

FIG. 20 is an address resolution table of the fifth embodiment, CA
It is a figure which shows an example of the content of M and a correspondence table.

FIG. 21 is a diagram showing a processing flow of the routing control unit of the fifth embodiment.

FIG. 22 is a diagram showing a configuration diagram of a routing control unit of the sixth embodiment.

FIG. 23 is a filtering table of the sixth embodiment, C
It is a figure which shows an example of AM, the content of a correspondence table, and the packet received by network interface 110A.

FIG. 24 is a diagram showing a processing flow of the routing control unit of the sixth embodiment.

[Explanation of symbols]

100 ... Router, 110 ... Network interface, 120 ... Filtering control unit, 130 ... Routing control unit, 140 ... Filtering table, 150
... cache memory, 151 ... CAM, 152 ... correspondence table, 154 ... routing table, 181-18
4 ... Network, 191-194 ... Terminals, 1010,
1020, 1410 to 1430 ... Packets, 1150 ...
Address resolution table

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yoshito Saiko 810 Shimoimaizumi, Ebina, Kanagawa Pref., Office Systems Division, Hitachi, Ltd.

Claims (12)

[Claims] 1. A plurality of networks to which a plurality of terminals are connected, and a plurality of network interfaces for mutually connecting these networks, each network interface including a routing table and a cache memory for searching the routing table. In a communication system including a router device having, the communication information between the terminals connected to different networks is received, the destination address is extracted from the received packet, and the routing table is searched using the destination address as a search key.
A method for searching a table in a communication system for relaying the information to another network, wherein a destination address from which a pointer to a corresponding entry on the routing table is obtained is C in the cache memory.
When registering in the AM and registering a pointer to the corresponding entry of the routing table in the correspondence table having an entry linked to the entry of the CAM and searching the routing table again for the same destination address, If the destination address is collated with the CAM and the corresponding address is registered, the CAM is
A method for searching a table, characterized in that a pointer to the routing table is obtained from an entry in the correspondence table linked to an entry in which the corresponding address is registered. 2. A communication system comprising a plurality of networks to which a plurality of terminals are connected and a plurality of network interfaces for connecting these networks to each other, and communication between terminals connected to different networks. A communication system which receives information, extracts a destination address from a received packet, searches a routing table of a router device in the network interface using the destination address as a search key, and relays the information to another network, The router device has a routing table and an address resolution table for mapping the address of the network layer, which is the third layer of the OSI reference model of ISO, and the address of the data link layer, which is the second layer, The routing table and In search of dresses resolution table, the destination address pointer is obtained to the appropriate entry on the table of the cache memory CAM
When registering to the CAM and registering a pointer to the corresponding entry of the above table in the corresponding table having an entry linked to the entry of the above CAM and searching the above table again for the same destination address, The destination address is collated, and if the corresponding address is registered, the pointer to the table is obtained from the entry of the correspondence table linked to the entry in which the corresponding address of the CAM is registered. Table search method. 3. A communication system comprising a plurality of networks to which a plurality of terminals are connected, and a plurality of network interfaces for connecting these networks to each other, and communication between the terminals connected to different networks. A communication system that receives information, extracts a destination address from a received packet, searches a routing table of a router device in the network interface using the destination address as a search key, and relays the information to another network, The router device has an address resolution table for mapping an address of a network layer, which is the third layer of the ISO OSI reference model, and an address of a data link layer, which is the second layer, and a bridge filtering table. , One of the above In search of Le, the correspondence table with an entry of the destination address pointer is obtained to the appropriate entry on the table registered in the cache memory CAM, linked to the entry of the CAM,
When registering a pointer to the corresponding entry in the table and searching the table again for the same destination address, the destination address is collated in the CAM, and if the corresponding address is registered, the CAM corresponds A table search method characterized in that a pointer to the table is obtained from an entry in the correspondence table linked to an entry in which an address is registered. 4. The table search method according to claim 1, 2 or 3, wherein the destination address is extracted from the received packet, the destination address is collated with the CAM, and it is determined whether or not the destination address is registered. , If there is registration, a pointer to the table is acquired from the entry of the corresponding table linked to the entry in which the corresponding destination address of the CAM is registered, and if there is no registration, the destination address is searched. A table search method characterized by directly searching the above table as a key. 5. The table search method according to claim 1, 2 or 3, wherein a destination address is extracted from the received packet, the table is directly searched using the destination address as a search key, and the destination address is stored. If a matching entry is found, the destination address is registered in the empty entry of the CAM, and a pointer to the corresponding entry in the table is registered in the entry in the correspondence table linked to the entry. Characteristic table search method. 6. The table search method according to claim 5, wherein the destination address, which is a search key for the table search, is larger than the size of the CAM entry.
A table search method using a hash function for changing the search key to the entry size of the CAM. 7. The table search method according to claim 6, wherein the destination address is extracted from the received packet, and the destination address is passed through the hash function for changing the search key to the entry size of the CAM, A table search method characterized in that the CAM is collated with the destination address passed through the hash function as a search key. 8. The table search method according to claim 5, wherein the CAM entry is used when the destination address and other information are used as a search key for the table search depending on the type of protocol. A table search method characterized in that the destination address and the other information are registered in. 9. A plurality of networks to which a plurality of terminals are connected, and a plurality of network interfaces for interconnecting these networks, each network interface having a router device having a routing table, and different networks. In a network device that receives communication information between terminals connected to each other, extracts a destination address from the received packet, searches the routing table of the router device using the destination address as a search key, and relays the information to another network. The router device has a CAM having an entry for registering a destination address where a pointer to the corresponding entry on the routing table is obtained in the cache memory, and an entry for registering a pointer to the corresponding entry in the routing table. When searching the routing table for a destination address and a cache memory having a correspondence table, the destination address is collated with the CAM, and if the relevant address is registered, the entry in which the relevant address of the CAM is registered And a routing control circuit for obtaining a pointer to the routing table from an entry of the correspondence table linked to the router device. 10. A network comprising a plurality of networks to which a plurality of terminals are connected and a plurality of network interfaces for connecting these networks to each other, and receiving communication information between terminals connected to different networks. In a communication system that extracts a destination address from a received packet, searches a routing table of a router device of a network interface using the destination address as a search key, and relays the information to another network, the router device includes a routing table, An address resolution table for mapping an address of a network layer, which is the third layer of the OSI reference model of ISO, and an address of a data link layer, which is the second layer, is provided. , On each table above A cache memory that has a CAM that registers a destination address from which a pointer to this entry is obtained, and a correspondence table that has an entry linked to the entry of the CAM and that has a correspondence table that registers a pointer to the corresponding entry in the above table; When searching the table for an address, the destination address is collated with the CAM, and if the corresponding address is registered, from the entry of the correspondence table linked to the entry in which the corresponding address of the CAM is registered. A router device comprising: a routing control circuit for obtaining a pointer to the table. 11. A communication system comprising a plurality of networks to which a plurality of terminals are connected, and a plurality of network interfaces for connecting these networks to each other, and communication between the terminals connected to different networks. A communication system which receives information, extracts a destination address from a received packet, searches a routing table of a router device in a network interface using the destination address as a search key, and relays the information to another network, wherein the router The device includes a routing table, an address resolution table for mapping an address of a network layer, which is the third layer of the ISO OSI reference model, and an address of a data link layer, which is the second layer, and a bridge filtering table. Equipped with , And corresponding entries in the table, each of which has a CAM that registers a destination address from which a pointer to the entry in the table is obtained and an entry linked to the entry of the CAM. A cache memory having a correspondence table for registering a pointer to, and searching the above table for the destination address,
First, the destination address is compared with the CAM, and if the corresponding address is registered, the pointer to the table is obtained from the entry of the correspondence table linked to the entry in which the corresponding address of the CAM is registered. A router device comprising a control circuit. 12. The router device according to claim 9, 10 or 11, wherein the CAM has a search key for searching the table, and the entry of the CAM and the entry of the correspondence table are one-to-one or A router device characterized by being linked to one to many.