JPH0669928A - Routing system - Google Patents

Abstract

PURPOSE:To reduce the time to retrieve an entry of revision object when routing information is received from a network and the routing table stored in an end system is revised. CONSTITUTION:A routing table is stored in a routing information storage section 12 and the revision processing is implemented by a routing information management section 13. In the cafe of making data transfer, a routing section 11 obtains a destination physical address from the routing table. A frequency of use information column is provided to each entry of the routing table and when the routing information including an address mask is received, entries whose use frequency information exceeds a threshold level are objects of retrieval and revision processing is performed. Or routing table is divided into a table including an address mask and a table not including the address mask, and when the routing information including the address mask is received, only the table including the address mask is retrieved and subject to revision processing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a routing system in a communication system in which a plurality of networks are connected to each other.

[0002]

2. Description of the Related Art FIG. 5 is a diagram showing an outline of a communication system in which a plurality of networks are mutually connected. In FIG. 5, 1, 2, 4, 5, 7, 8 are end systems such as workstations, 3 and 6 are relay systems, 10, 20,
30 is a network. In such a system, a physical address is used to specify the destination of data transfer between systems connected to the same network.
However, if the program specifies where to transfer the data,
Generally, a network address (a common address in a plurality of networks) is used, and therefore it is necessary to convert the network address into a physical address when executing the transfer. Therefore, each end system holds a routing table indicating the correspondence between network addresses and physical addresses in the routing information storage unit.

FIG. 6 is a block diagram showing an outline of a routing-related part in the end system. In FIG. 6, 11 is a routing unit, 12 is a routing information storage unit, 13 is a routing information management unit, and 14 is a data transmission / reception unit. The routing unit 11 obtains the physical address of the transfer destination from the routing table of the routing information storage unit 12 for data transfer, and gives it to the data transmitting / receiving unit 14. The contents of the routing table are
It is updated based on the routing information sent to the end system regularly or irregularly. The management is performed by the routing information management unit 13.

The routing information sent to the end system is mainly as follows. (1) Routing information periodically sent from the relay system connected to the network. This is information for the relay system to indicate to each end system in the network to which it is connected that it is currently in this network. The routing information includes the network address and physical address of the relay system.

(2) Routing information sent aperiodically from a relay system connected to the network. There are the following cases when this is sent. For example, when transferring data from the end system 1 on the network 10 to the end system 7 on the network 30 in FIG. 5, first, the data is transferred from the end system 1 to the relay system 6, and then from the relay system 6 to the end system 7. Will be transferred to. However, when the network address of the end system 7 is not registered in the routing table of the end system 1 and it is unknown to which physical address the address should be sent, it cannot be exchanged and any relay system, for example, the closest one is located. Send to the relay system 3. Then, the relay system 3 transfers the data to the relay system 6 which should be originally sent based on the routing information managed by the relay system 3 and correct routing information when transferring the data to the end system 1 to the end system 1. To send. The routing information includes the network address of the end system 7, the address mask, and the physical address of the relay system 6.

As described above, the routing information transmitted from the network side to the end system includes those having no address mask and those having no address mask. The former has a one-to-one correspondence between network addresses and physical addresses, and the latter has a plurality of network addresses and one physical address.

Here, the address mask will be described. The address mask makes it possible to point to a plurality of network addresses by taking a bit logical product of the address and the network address. For example, when the bit logical product of the network address 01111101101100001 (binary notation; the same applies below) with the address mask 11111111100000000 is taken, only the part of the network address where the bit of the address mask is 1 is valid, Indicates all network addresses having the upper 8 bits.

On the other hand, in a communication system in which a plurality of networks are connected to each other, in the network address of each end system, some of the upper digits are numbers indicating the network, and the remaining lower digits are numbers indicating the end systems. Has become. Therefore, if you match the part where the bit of the address mask is 1 with the number of digits of the network, you can collectively show all end systems on the same network by taking the bit logical product of that and the network address. You can

FIG. 7 is a diagram showing an example of a routing table in the conventional routing method. Each entry of the routing table holds a network address of each end system or relay system, a corresponding physical address, and a timer.
The timer is set to a predetermined value when the routing information about the network address is sent, and decreases with the passage of time until the next routing information is sent. Then, when the value of the timer becomes 0, the end system having the network address is regarded as disconnected from the network and is deleted from the routing table.

When the end system receives the routing information and the address mask is not included in the routing information, a new entry is registered in the routing table or an already registered entry is updated.
On the other hand, if the address mask is included, the result of the bit logical product of the received network address and the address mask and the result of the bit logical product of each network address of the routing table and the received address mask are calculated. Compare and update the entry if they match.

[0011]

(Problems) However, the above-mentioned conventional technique has the following problems. The first problem is that it takes time to search for a matching network address when the routing information including the address mask is received and the routing table is updated.

The second problem is that the routing information including the address mask does not make effective use of the physical addresses corresponding to a plurality of network addresses, even though they indicate the physical addresses.

(Description of Problems) First, the first problem will be described. In the conventional routing method, when the routing information including the address mask is received, the entire routing table is scanned and the bit logical product of the received address mask and each network address in the table is taken. Then, it is checked whether or not the result matches the bit logical product of the received network address and the address mask. On the other hand, although many large-scale networks have been made in recent years, the size of the routing table also increases as the scale of the network increases. Therefore, the amount of calculation for obtaining the bit logical product increases, and the search takes time. If the search takes a long time, the CPU in the end system is tied up by that much, the other processing is delayed, and eventually the entire processing is delayed.

Next, the second problem will be described. In the conventional routing method, even when the routing information with the address mask is received, only the network address and the corresponding physical address are registered in the routing table. Therefore, although the routing information including the address mask indicates the physical addresses corresponding to multiple network addresses, when data is transferred, when the routing table is searched for the destination physical address, the routing information is The physical address can be given only to the network address registered in the table. An object of the present invention is to solve the above problems.

[0015]

In order to solve the above problems, in the first invention of the present case, in the routing system of a communication system in which a plurality of networks are mutually connected, each entry of a routing table held by an end system is A usage frequency information column is provided, and when the end system receives the routing information including the address mask, only the entry whose usage frequency information exceeds the threshold value is searched and the update processing is performed. Further, according to the second aspect of the present invention, in a routing system of a communication system in which a plurality of networks are connected to each other, a first routing table storing routing information not including an address mask and an address mask are used as a routing table of an end system. And a second routing table for storing the routing information including the address information. When the end system receives the routing information including the address mask, only the second routing table is searched for the update process to transfer the data. When performing the above, first, the network address of the second routing table is obtained by referring to the network address of the first routing table, and when the destination is not found in the first routing table. And ad It was to determine the destination physical address with reference to the Sumasuku.

[0016]

[Operation] In the first invention, a usage frequency information column is provided for each entry of the routing table held in the end system, and when the end system receives the routing information including the address mask, the above-mentioned usage frequency information is provided. Update is performed by searching only for the entries that exceed the threshold. Therefore, since the number of entries to be searched is limited, the search time can be shortened. Also,
In the second invention, as the routing table of the end system, a first routing table that stores routing information that does not include an address mask and a second routing table that stores routing information that includes an address mask are provided. When the routing information including the address mask is received by the
Updates by searching only the routing table of. Therefore, since the number of entries to be searched is limited, the search time can be shortened. Further, at the time of data transfer, the destination physical address is obtained by referring to the network address of the first routing table, and when the destination is not found in the first routing table, the second routing table of the second routing table is searched. The destination physical address is determined by referring to the network address and the address mask. Therefore, a set of network addresses and address masks can cover up to other end systems on the network, and routing information including address masks indicating physical addresses corresponding to a plurality of network addresses. Can be effectively utilized.

[0017]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a diagram showing an example of a routing table applied to the first invention. This is the addition of usage frequency information for each entry and a flag indicating that the usage frequency has exceeded a threshold value to the conventional routing table. The usage frequency information indicates how many times the entry has been used per fixed time during data transfer, and is incremented by 1 each time the routing unit 11 (see FIG. 6) uses the entry in the table once. . Further, the flag is turned on when the usage frequency exceeds a threshold value.

FIG. 2 is a flow chart showing a processing procedure when receiving the routing information in the first invention. Step 1 ... It is determined whether the received routing information includes an address mask. Step 2 ... When the address mask is included, it is determined whether or not the flag of the routing table is 1. Step 3 ... When the flag is 1, the bit logical product of the network address of the entry and the address mask included in the routing information is obtained, and the value obtained as a result and the network address included in the routing information are obtained. The value obtained by taking the bit logical product of the address mask is collated to determine whether they match. Step 4 ... When they match, the timer of the entry is reset to the initial value. When the physical address included in the routing information is different from the physical address of the entry, the physical address of the entry is replaced with the physical address included in the routing information. Step 5 ... If they do not match, move to the next entry. Step 6 ... It is determined whether or not all the entries in the routing table have been searched. Step 7 ... When all the entries have been searched, the network address and physical address included in the routing information are registered in the routing table.

Step 8 ... When the address mask is not included in Step 1, the network address of the entry is collated with the network address included in the routing information to determine whether they match. Step 9 ... When they match, the timer of the entry is reset to the initial value. When the physical address included in the routing information is different from the physical address of the entry, the physical address of the entry is replaced with the physical address included in the routing information. Step 10 ... If they do not match, move to the next entry. Step 11 ... It is determined whether or not all the entries in the routing table have been searched. Step 12 ... When all the entries have been searched, the network address and physical address included in the routing information are registered in the routing table.

With this configuration, when the routing information including the address mask is received and the routing table is updated, only the frequently used entry is searched, so that the search time can be saved. On the other hand, an entry with a low frequency of use is not updated even when routing information comes in, and even if it is registered in the table, it will be deleted from the table when the timer value becomes 0, but the data will be transferred there. In that case, if you send it to one of the relay systems, that relay system will find the correct physical address and transfer it. At that time, it can be registered in the table again based on the routing information returned from the relay system. Transferring in this way takes more time than transferring by obtaining a physical address from the routing table, but since the transfer frequency is originally low, there is no problem.

Next, the second invention will be described. Figure 3
FIG. 6 is a diagram showing an example of a routing table applied to the second invention. In the present invention, the routing table is divided into two. FIG. 3A shows a first routing table that stores routing information that does not include an address mask, and FIG. 3B shows a second routing table that stores routing information that includes an address mask. ing.

FIG. 4 is a flow chart showing a processing procedure when receiving the routing information in the second invention. Step 1 ... It is determined whether the received routing information includes an address mask. Step 2 ... When the address mask is included, the bit logical product of the network address of the entry and the address mask is obtained, and the bit logical product of the value obtained as a result and the network address included in the routing information and the address mask Is compared with the value obtained by taking to determine whether or not they match. Step 3 ... When they match, the timer of the entry is reset to the initial value. When the physical address included in the routing information is different from the physical address of the entry, the physical address of the entry is replaced with the physical address included in the routing information. Step 4 ... If they do not match, move to the next entry. Step 5 ... It is determined whether or not all the entries in the routing table have been searched. Step 6 ... When all the entries have been searched, the network address, address mask and physical address included in the routing information are registered in the second routing table.

Step 7 ... When the address mask is not included in Step 1, the network address of the entry is collated with the network address included in the routing information to determine whether they match. Step 8 ... When they match, the timer of the entry is reset to the initial value. When the physical address included in the routing information is different from the physical address of the entry, the physical address of the entry is replaced with the physical address included in the routing information. Step 9 ... If they do not match, move to the next entry. Step 10 ... It is determined whether or not all the entries in the routing table have been searched. Step 11 ... The first when all the entries have been searched
The network address and the physical address included in the routing information are registered in the routing table of.

The routing unit 11 (see FIG. 6) performs the following processing when transferring data. (1) The network address of the data transfer destination is sequentially compared with the network address of each entry of the first routing table, and if there is an entry having a matching network address, the physical address of that entry is used as the data transmitting / receiving unit 14 Send to. (2) If no matching entry is found in (1), the network address of the data transfer destination and the second
The value obtained by taking the bit-logical product of the address mask of each entry of the routing table and the value obtained by taking the bit-logical product of the network address of the entry and the address mask are sequentially collated. As a result, if there is a matching entry, the physical address of the entry is sent to the data transmitting / receiving unit 14.

[0025]

As described above, the routing method of the present invention has the following effects. (1) In the first invention, a usage frequency information column is provided for each entry of the routing table held by the end system, and when the end system receives the routing information including the address mask, the usage frequency information is Only the entries that exceed the threshold are searched and updated. Therefore, since the number of entries to be searched is limited, the search time can be shortened. (2) In the second invention, a first routing table that stores routing information that does not include an address mask, as a routing table of the end system,
Second, storing routing information including address mask
When the end system receives the routing information including the address mask, only the second routing table is searched and the update processing is performed. Therefore, since the number of entries to be searched is limited, the search time can be shortened. Further, at the time of data transfer, the destination physical address is obtained by referring to the network address of the first routing table, and when the destination is not found in the first routing table, the second routing table of the second routing table is searched. The destination physical address is determined by referring to the network address and the address mask. Therefore, a set of network addresses and address masks can cover up to other end systems on the network, and routing information including address masks indicating physical addresses corresponding to a plurality of network addresses. Can be effectively utilized.

[Brief description of drawings]

FIG. 1 is a diagram showing an example of a routing table applied to the first invention.

FIG. 2 is a flowchart showing a processing procedure when receiving routing information in the first invention.

FIG. 3 is a diagram showing an example of a routing table applied to the second invention.

FIG. 4 is a flowchart showing a processing procedure when receiving routing information in the second invention.

FIG. 5 is a diagram showing an outline of a communication system in which a plurality of networks are mutually connected.

FIG. 6 is a block diagram showing an outline of a routing-related part in the end system.

FIG. 7 is a diagram showing an example of a routing table in a conventional routing method.

[Explanation of symbols]

1, 2, 4, 5, 7, 8 ... End system, 3, 6 ... Relay system, 10, 20, 30 ... Network, 11 ...
Routing unit, 12 ... Routing information storage unit, 13
... Routing information management unit, 14 ... Data transmission / reception unit

Claims (2)

[Claims] 1. In a routing system of a communication system in which a plurality of networks are mutually connected, a usage frequency information column is provided for each entry of a routing table held in an end system, and the end system includes a routing including an address mask. A routing method characterized in that, when the information is received, the update processing is performed by searching only for the entry in which the usage frequency information exceeds a threshold value. 2. In a routing system of a communication system in which a plurality of networks are connected to each other, as a routing table of an end system, a first routing table that stores routing information that does not include an address mask, and routing information that includes an address mask. And a second routing table for storing data, and when the end system receives the routing information including the address mask, only the second routing table is searched for update processing, and data is transferred. Has
First, the destination physical address is obtained by referring to the network address in the first routing table, and if the destination is not found in the first routing table, the network address and the address mask in the second routing table are referred to. The routing method is characterized in that the destination physical address is determined.