JPH06332829A - Network connection method - Google Patents

Abstract

(57) [Summary] [Purpose] To enable connection to a subnet without changing the processing unit of the network layer of the computer. [Structure] The address translation table 8 of the address translator 1 stores the logical address and physical address pair of another computer on the physical network to which the computer is directly connected. The subnet configuration information table 9 stores its own address. A logical address of a computer on another subnet to which the computer is not directly connected and a physical address of a router relaying when communicating with the computer are stored. The network layer processor 2 gives the logical address of the computer of the communication partner to the address translator 1. The address translation device 1 has a table 8,
9, the corresponding physical address is output.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of connecting an electronic computer system to a network in which one logical network is composed of a plurality of physical networks.

[0002]

2. Description of the Related Art Conventionally, regarding a network configuration, there has been a one-to-one correspondence between a logical network and a physical network. With the spread of LANs (local area networks), large-scale networks in which many physical networks are connected to each other have come into use. As a result, in such a large-scale network, management information about the logical network increases in order to associate the logical network and the physical network one-to-one, and many resources for management (memory, disk, etc.) Equipment).

Therefore, in order to avoid this, a technique is used in which management information on a logical network is reduced by associating a logical network with a physical network in a one-to-many relationship. This technology is called "subnet".

In the subnet, there is inconsistency between the logical information about the network and the physical reality (especially the information of the relay route). Therefore, the process for correcting the inconsistency occurs in the network layer portion of the communication protocol. You will need it. This is because the network layer has the function of selecting an appropriate repeater (hereinafter referred to as "router") when communicating with a logical network that is not directly connected to the computer, but in the subnet, it is physically Is not directly connected, but when communicating to a computer of a network that logically belongs to the same network, it cannot be correctly communicated because it is determined that the original processing of the network layer does not need to go through a router. Is.

Therefore, in order to connect a computer to the subnet, the network layer processing device of each computer must be changed. Conversely, alternative means must be used to connect the unmodified computer to the subnet. For example, by causing a router to execute a special protocol and responding to an inquiry issued from an address resolution device of a computer, the router responds by acting to trick the address resolution device to realize a subnet connection. This protocol is called Proxy ARP (proxy address resolution protocol).

[0006]

In the method of changing the network layer processing device of the existing computer, since the processing of the network layer is generally complicated, there are problems in the cost of the change and the reliability after the change. In addition, P
The method of executing roxy ARP has a problem that the load of the router becomes high and it becomes difficult to solve the trouble when it occurs.

An object of the present invention is to provide a network connection system which enables connection to a subnet without changing the processing device of the network layer of a computer.

[0008]

In order to achieve the above object, according to the present invention, one logical network is composed of a plurality of physical networks, and information processing devices are mutually connected via the physical networks. In the connection method, the first information processing apparatus directly connected to the first
A first table storing a correspondence relationship between a logical address and a physical address of a second information processing device on the second physical network, and a second table on the second physical network not directly connected to the first information processing device. No. 3 information processing apparatus, and a second table storing the correspondence between the logical address of the third information processing apparatus and the physical address of the router relayed to the third information processing apparatus, and the first information processing apparatus and the second table. When communication is performed between the third information processing devices, the network layer processing means in the first information processing device gives the logical address of the information processing device of the communication partner to the table searching means, and the table searching device Referring to the first and second tables, a physical address corresponding to the logical address of the information processing apparatus of the communication partner is generated, and the network layer processing means generates the physical address. As a destination, it calls the communication device control means is characterized by executing the data transmission.

[0009]

The address translation device of the computer is called from the processing device of the network layer, and the logical address (address of the protocol upper layer such as network layer and transport layer) of the communicating party is used as the physical address (depending on the actual transmission medium) Address) and a mechanism for converting the address.
The address translation device usually consists of a simple table search algorithm and an inquiry protocol process when it cannot be translated. A device that stores the subnet configuration information is added to this address translation mechanism, and the subnet configuration information, that is, the logical address information that determines each accessible subnet, and the router that relays to access the subnet. Store the physical address.

In an ordinary algorithm of an address translation device, an address translation table in which a logical address of a computer and a physical address are associated with each other is searched, and if found, the physical address is associated with the address. Attempt to solve by the inquiry protocol by the simultaneous broadcast. On the other hand, in the present invention, when the address translation table cannot be found by searching, the router for accessing the subnet based on the stored configuration information of the subnet before attempting to solve the error reporting or inquiry protocol. Discover the physical address of
If it can be found, the physical address is associated with the physical address corresponding to the logical address of the communicating party. As a result, the processing device of the network layer that called the address translation device does not have the information about the configuration of the subnet and does not have the special function for the subnet, but the logical address of the communication partner and the router It succeeds in associating a physical address. The algorithm of the address translator is generally simple, and the processing newly added to it can be realized by a simple search algorithm.
It is easy to change and the reliability of the entire system is not compromised.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be specifically described below with reference to the drawings. First, conventional network layer processing will be briefly described. The processing of the network layer is route selection, that is, determining the logical address of the router through which the target machine is communicated from the logical address of the target machine given as the destination. If the target machines are on the same network, no router is required, so there is virtually no processing (of course, processing other than route control, such as data division, is performed).

The network layer internally has a table, which is called route information, which is a pair of a logical network to which a target machine belongs and a logical address of a router required to communicate therewith. Then, the route selection process described above is performed using this table.

The processing flow of the network layer is as follows: (1) Called from the upper transport layer, and (2) referring to the route information, and communicating from the logical address of the target machine given as the destination to the target machine. Determine the logical address of the router through which it does (this process is not done if the target machine is on the same network),
(3) Pass the logical address of the router to the address translator (if the target machine is on the same network, pass the logical address of the machine itself), and (4) communicate with the physical address returned from the address translator as the destination. Call the device control unit and execute data transmission. At this time, the process of the address translator determines the physical address corresponding to the logical address of the target machine existing in the same logical network.

In this case, the network layer is "one logical network corresponds to one physical network".
If it works on the assumption that it does not work properly under the subnet environment. FIG. 2 shows a configuration example of a network. The computer 12 (A) and the computer 13 (C) are connected to the physical network 10a, the computer 14 (B) is connected to the physical network 10b, and the physical network 10 is connected.
A and 10b are connected by a router (relay computer) 15. And the two physical networks 10
A logical network 11 is formed by a and 10b.

In FIG. 2, since the computer B belongs to the same logical network as its own computer A, it is determined that no router is interposed between the computer B and the computer B, so that it does not operate properly in the subnet environment.

Therefore, in order to operate normally in the subnet environment, the network layer abandons the premise that one logical network corresponds to one physical network, and the target machines belonging to the same logical network are abandoned. Even for transmission, processing must be performed on the assumption that the router may intervene. In other words, it is necessary to perform processing that considers the subnet configuration.

Of course, it is possible to change the network layer so that such processing can be performed, and there are many computers that implement such processing. However, in the present invention, there is no need for the network layer to execute complicated processing. It allows you to connect to a subnet without making any changes.

FIG. 1 is a block diagram of a computer equipped with the address translation device of the present invention. In the network of FIG. 2, for example, the configuration of the computer 12 (A) is as shown in FIG. In FIG. 1, the address translation device 1 is connected to the network layer processing device 2 and the communication device control device 3 by a bus or the like.

A logical address of a destination requested to be transmitted from an upper layer (a protocol upper layer such as a transport layer) is input to the bus 4 input from the network layer processing device 2 to the address translation device 1, and the network is also transmitted. A physical address (address corresponding to an actual transmission medium) corresponding to the logical address is output to the output bus 5 to the layer processing device 2.

The address translation device 1 comprises a table search device 6, an inquiry processing device 7, an address translation table 8 and a subnet configuration information table 9. The address conversion table 8 stores a pair of a logical address and a physical address of another computer on the physical network to which the own computer is directly connected. In addition, the subnet configuration information table 9 stores a list of logical addresses of computers on other subnets to which the own computer is not directly connected, and a physical address of a router that relays when communicating there.

When the logical address is input from the input bus 4, the address translation device 1 searches the address translation table 8 for a physical address corresponding to the logical address.
When the physical address corresponding to the logical address is found, the physical address is output to the output bus 5. If there is no corresponding physical address, the subnet configuration information table 9 is searched, and the physical address of the corresponding router is output to the output bus 5. If the physical address of the corresponding router is not found even when the table 9 is searched, the inquiry processing device 7 makes a transmission request to the communication device control device 3 in order to communicate the inquiry packet.

The address translation device will be described more specifically with reference to FIG. In the network configuration shown in FIG. 2, for example, when the computer A communicates with the computer B, it is necessary to relay the router R. The logical addresses of computers A, B, C and router R are A, B, C, R respectively,
When the corresponding physical addresses are a, b, c, and r, the address conversion table 8 of the computer A stores logical address and physical address pairs as shown in FIG.

Further, in the subnet configuration information table 9 of the computer A, as shown in FIG. 4, when the computer B communicates with the logical address of the computer B on another subnet which is not directly connected to the computer A itself. Router 15 to relay
The physical address r of (R) is stored.

The operation of the embodiment in the case of communicating from the computer A to the computer B will be described below. The address translation device 1 of the computer A receives the logical address B from the network layer processing device 2. The table search device 6 searches the address conversion table 8 to search for a physical address corresponding to the logical address B. Since the logical address B is not found as a result of the search, the subnet configuration information table 9 is searched next. Then, the logical address B is found, and the physical address r of the router R corresponding thereto is output to the network layer processing device 2. Then, the network layer processing device 2 calls the communication device control device 3 with the physical address r of the router R returned from the address translation device 1 as a destination, and performs data transmission.

As described above, according to the present invention, since the network layer is not changed, that is, the configuration of the subnet is not considered, it is judged that the transmission to the target machine belonging to the same logical network does not always go through the router, Pass the logical address of the target machine itself to the address translator. Then, in the address translation device, the configuration of the subnet is considered, the required router is determined,
The physical address of that router is returned to the network layer. From the network layer's point of view, the physical address of the target machine seems to have been returned, but the physical address of the router is actually being returned. Also,
Since the method of the present invention can be realized by changing the address translation device, the hardware cost is low.

[0026]

As described above, according to the present invention, the correspondence between the logical address and the physical address of the second information processing device on the first physical network directly connected to the first information processing device. A first table that stores the relationship, a logical address of a third information processing device on a second physical network that is not directly connected to the first information processing device, and a relay to the third information processing device A second table storing the correspondence with the physical address of the router
When communication is performed between the first information processing device and the second or third information processing device, the network layer processing means in the first information processing device uses the logical address of the information processing device of the communication partner as the table search means. The table search means refers to the first and second tables to generate a physical address corresponding to the logical address of the information processing apparatus of the communication partner, and the network layer processing means sends the generated physical address to the destination. As described above, since it is configured to call the communication device control means and execute data transmission, it is possible to connect to the subnet without making any changes to the network layer having a large function.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a computer including an address translation device of the present invention.

FIG. 2 shows a configuration example of a network.

FIG. 3 is a diagram showing the contents of an address conversion table.

FIG. 4 is a diagram showing the contents of a subnet configuration information table.

[Explanation of symbols]

 1 address conversion device 2 network layer processing device 3 communication device control device 4 input bus 5 output bus 6 table search device 7 inquiry processing device 8 address conversion table 9 subnet configuration information table

Claims (1)

[Claims] 1. In a network connection system in which one logical network is composed of a plurality of physical networks, and information processing apparatuses are mutually connected via the physical networks, a first information processing apparatus is directly connected to the first information processing apparatus. A first table storing a correspondence relationship between a logical address and a physical address of a second information processing device on one physical network; and a second table not directly connected to the first information processing device.
A second table storing a correspondence relationship between a logical address of the third information processing device on the physical network and a physical address of a router relaying to the third information processing device. When communication is performed between the processing device and the second or third information processing device, the network layer processing means in the first information processing device gives the logical address of the information processing device of the communication partner to the table search means, The table search means refers to the first and second tables to generate a physical address corresponding to the logical address of the information processing apparatus of the communication partner, and the network layer processing means uses the generated physical address as a destination. , A network connection method characterized by calling communication device control means and executing data transmission.