JPH05153148A - Routing system and communication controller - Google Patents

Abstract

PURPOSE:To easily cope with the increase or decrease of nodes, and to decide a direction by a simple algorithm by adding a 2-ary node address, in a network constituted of plural nodes branched into less than 3 directions. CONSTITUTION:A network 1 is expressed by a tree structure, and the node address of an (n)-order stage node is expressed by an (n) digit 2-ray. Then, at the time of communication, the node address of destination is compared with the node address of its own node within the range of the number of digits. When they disagree, the information is transferred to an upper node. When they are matched, the information is transferred to the left side when the node address of the destination at the left side by one digit than the digit of the node address of its own node is '1', and the information is transferred to the right side when the node address of the destination at the left side by one digit than the digit of the node address of its own node is '0'. Afterwards, the same operation is repeated, and when the node address of its own node is matched with the node address of the destination, the information is fetched.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a routing system for relaying a communication message and delivering it to a destination node in a network composed of a plurality of nodes branched within three directions.

With the progress of telecommunication technology, communication networks have become complicated and large-scaled. In such a communication network, when the number of nodes is increased or decreased, the operation engineer of the network can easily cope with it, and a simple routing method is required for the route determination algorithm at the time of communication.

[0003]

2. Description of the Related Art FIG. 5 is a diagram for explaining a conventional communication control device for a node. In the figure, 10, 21 and 22 are ports,
11, 211 and 221 are transmission buffers (shown as S in the figure), 12, 212 and 222 are reception buffers (shown as R in the figure), 30 is a microprocessor (hereinafter referred to as MPU), and 31 to 33 are direct memory access. A controller (hereinafter referred to as DMAC), 41 is a node address management memory (shown as a node ADDR management memory in the figure).

In the figure, port 10 is connected to an upper node (not shown), port 21 is connected to a lower node on the left (not shown), and port 22 is connected to a lower node on the right (not shown). It is connected.

In such a configuration, when one node communicates with an arbitrary node, the source node sends the destination node address, and the MPU 30 of the node receiving the node address stores the contents of the node address management memory 41. , The node connection state is confirmed to determine the routing, and the information is transmitted in the designated direction via any one of the ports 10, 21, and 22.

In order to speed up the transfer of information, the DMAC3
With the control of 1 to 33, the transmission buffers 11, 211, 22
1, the MPU 30 between the reception buffers 12, 212, 222
Information is directly transmitted and received without going through.

In addition, when the number of nodes is increased or decreased in such a network, the network operation engineer determines the node address of the new node after confirming the network configuration and the node address of each node at that time. ..

Since each node has a node address management memory 41 to determine the routing for communication, when the number of nodes is increased or decreased, the node address management memories 41 of all the nodes in the network are used. It will be necessary to write the node address of the new node.

[0009]

In the above-mentioned conventional example, when a node is added, the network operation engineer confirms the network configuration and the node address of each node at that time, and determines the numbering system. Determine a node address that does not conflict with.

Therefore, setting the node address of the new node is a complicated task, and adding the node address of the new node to all the nodes in the network requires a long time.

Further, the operation of route setting in the node at the time of communication is complicated. The present invention intends to realize a routing method that can easily cope with an increase or decrease in the number of nodes and can determine a route with a simple algorithm.

[0012]

FIG. 1 is a diagram for explaining the principle of the present invention. The figure shows the network 1 in a tree structure. The binary number "0" is the node address of the node on the highest ladder, and the binary number "10" is the node address of the node on the left of the next ladder. Yes, "00 is the node address of the node on the right side of the next rank.

In the following, the node address of the node of the nth rank is an n-digit binary number, and n- of the nodes of the nth rank is
The node address of the node connected to the left side of the primary node is “1” added to the left digit of the n−1th order node address, and the node address of the node connected to the right side of the n−1th order node is "0" is added to the left digit of the (n-1) th order node number.

At the time of communication, the destination node address is compared with the range of the number of digits of its own node address, and if they do not match, the information is transferred to the upper node. When the destination node address of is "1", the information is transferred to the left side, and the destination node address one left of the digit of the own node address is "0".
In the case of, the information is transferred to the right side, and thereafter, the same operation is repeated, and when the own node address and the destination node address match, the information is fetched.

[0015]

The network 1 of a form in which one node branches within three directions is represented by a tree structure, the node address of the node at the highest rank is 0, and the node address of the node on the left side of the next rank is 10 And the node address of the node on the right side of the next level is 00.

Similarly, the node address of the node of the nth rank is represented by an n-digit binary number, and the node address of the node connected to the left side of the node of the n-1th rank among the nodes of the nth rank. Assigns “1” to the left digit of the n−1th order node address, and the node address of the node connected to the right side of the n−1th order node is “0” to the left digit of the n−1th order node address. The node address is autonomously assigned by adding

Therefore, when the number of nodes increases or decreases,
The node address design by the operation engineer is unnecessary. Also, during communication, the destination node address is compared with the range of the number of digits of its own node address, and if they do not match, they are transferred to the upper node. Information is transferred to the left when the node address of the node is "1", and when the destination node address one left of the node address of the node is "0", the information is transferred to the right. The connection operation can be performed by an algorithm.

Therefore, the configuration of the communication control device of the node can be simplified.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 2 is a diagram for explaining a network to which the present invention is applied. When the direction of branching from one node is three or less, the network can be represented as a tree structure.

FIG. 1 shows the network of FIG. 2 as a tree structure. Here, the node address of each node is given as shown in the figure, and the node address can be uniquely and autonomously determined. However, only the root node 0 branches in only one direction or two directions.

FIG. 3 shows a flow chart of routing according to the embodiment of the present invention. In the figure, node address 011
A flow chart will be described in the case of communicating from the node of 10 to the node of the node address 100.

The start of the connection operation is shown. The own node address 01110 and the destination node address 100 are compared. Since the comparison result does not match, the process proceeds to.

Own node address 01110 (5 digits)
And the destination node address 100 (3 digits),
Based on the number of digits of the local node address, comparison is performed within that range. Again, there is no match, so proceed to.

Information is transferred to the upper node 1110. At this point, the upper node to which the information is sent becomes the own node. In the upper node 1110, the operation of is repeated, but they do not match.
Transfer to 0 and 10 and perform the same operation, but they do not match.

Similarly, the information arrives at node 0, where the first digit is the same, so the procedure advances to. It is confirmed whether or not the digit to the left of the first digit of the destination node address is 1.

When the comparison result is 1, the information is transferred to the left side. Here, since the leftmost digit of the rightmost "0" of the destination node address is "0", the information is transferred to the right side. Here, the process proceeds to node 00.

Here, the procedure returns to and the same processing is performed. Here, since the digit to the left of 00 is 1, the information is transferred to the node 100 on the left side and the process returns to. Then, since the destination node address 100 and the own node address 100 match, the information sent is the own node 1
It is judged that the information is addressed to 00 and is taken into the own node 100, indicating the end.

In this way, each node can perform connection processing with a simple algorithm that compares its own node address with the destination address and determines the direction of information transmission based on the comparison result.

FIG. 4 is a diagram for explaining the communication control device of the node according to the embodiment of the present invention. Ports 10, 21, 2 in the figure
2, transmission buffers 11, 211, 221, reception buffers 12, 212, 222, MPU 30, DMACs 31 to 3
3 is the same structure as described in the conventional example.

Reference numeral 40 is a node address memory in which only the own node address is written. In the conventional example, the node addresses of all the nodes in the network are written in the node address management memory 41 of each node,
Although it is necessary to always manage the latest address, in the embodiment of the present invention, it is only necessary to write the own node address.

In this way, the configuration of the communication control device of the node can be simplified.

[0032]

According to the present invention, communication is performed by giving a binary node address that is uniquely determined depending on whether the node is connected to the left or right side of the ladder and the upper node. When doing the routing,
A routing method that can apply a common algorithm to all nodes can be realized. Further, since the node address is autonomously determined, it is not necessary for the operation engineer to design the node address even if the number of nodes increases or decreases.

Further, since the routing can be performed only by the own node address and the destination address and the memory amount for writing the node address is reduced, the communication control device can be simplified.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a diagram illustrating a network that is a target of the present invention.

FIG. 3 is a flowchart of routing according to an embodiment of the present invention.

FIG. 4 is a diagram illustrating a node communication control device according to the embodiment of this invention.

FIG. 5 is a diagram illustrating a communication control device of a node of a conventional example.

[Explanation of symbols]

 1 network 0-0000, 10-1110, 01110 node 10, 21, 22 port 11, 211, 221 transmission buffer 12, 212, 222 reception buffer 30 MPU 31-33 DMAC 40 node address memory 41 node address management memory

Claims (2)

[Claims] 1. A routing method for a network (1) that branches from one node within three directions, wherein the network (1) is represented by a tree structure, and the node address of the node at the highest rank is It is a 1-digit binary number "0", and the node address of the node on the left side of the next rank is 2-digit 2
Decimal number is “10” and the node address of the right node is 2
It is a binary number "00", and the node address of the node of the nth rank is an n digit binary number, and the node address of the node connected to the left side of the node of the n-1st rank among the nodes of the nth rank Assigns “1” to the left digit of the n−1th order node address, and the node address of the node connected to the right side of the n−1th order node is “0” to the left digit of the n−1th order node address. When communicating, the destination node address is compared with the range of the number of digits of its own node address. If they do not match, the information is transferred to the upper node, and if they match, 1 from the digit of their own node address. If the destination node address on the left side is "1", the information is transferred to the left side, and the value is 1 from the digit of your node address.
When the destination node address on the left side is "0", the information is transferred to the right side, and the information is fetched when the own node address and the destination node address match. 2. A port (10) for connecting to an upper node
Write the port (21) connected to the lower node on the left side, the port (22) connected to the lower node on the right side, the control device (30) that controls connection and communication processing, and your own node address. A communication control device comprising a node address memory (40).