JPH07235945A - Communication control system for ring bus system - Google Patents

Abstract

PURPOSE:To prevent deterioration of the processing capability despite the congestion of a ring bus and also to secure the stable processing capability regardless of the traffic state in a ring bus system. CONSTITUTION:A communication management node l is provided with a normal token generating/monitoring means 11 which generates a normal token frame that circulates on a ring bus and also monitors the circulating time on the ring bus. A load state monitoring control means 12 decides whether the circulating time acquired by monitoring the normal token frame continuously exceeds the prescribed time. If decided so, an individual token generating/monitoring means 13 generates an individual token frame added with priority in addition to the normal token frame after deciding a congestion state of the ring bus. Then the means 13 sends the individual token frame onto the ring bus to monitor it.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a communication control system in a ring bus system adopting a token passing system.

In a ring bus system in which a plurality of communication nodes having a data transmission / reception function are connected to a ring-shaped bus, a communication management node that manages at least one system is provided, and one token frame circulating on the ring bus is provided. A token passing method is known in which the captured communication node obtains the data transmission right.

When this token passing system is used, if a large number of communication nodes in the ring bus request communication, the cycle time of the token frame becomes long and the processing capacity of the system is lowered, so that improvement is made. Is desired.

[0004]

2. Description of the Related Art FIG. 7 is an explanatory diagram of a ring bus system, and FIG. 8 is a configuration example of a communication node. A. of FIG. Is the configuration of the ring bus system, B. Is a frame structure.

Referring to FIG. , 70 to 75 are general communication nodes (displayed as # 1 to # 6), 76 is a communication management node having a function of controlling a plurality of communication nodes and a ring bus (displayed at # 7), and RB0 and RB1 are each Represents a ring bus for connecting communication nodes. In this configuration, two ring buses RB0 and RB1 are provided, but one (for example, RB0) is currently used and the other (for example, RB1)
Is used as a backup when the current one is in failure. Incidentally, the ring bus provided in the present number n is called a ring bus having an n + 1 configuration. In this ring bus system, a communication node that captures one token circulating on the ring bus obtains a data transmission right and transmits the transmission data to another communication node.

FIG. 7B. A frame structure used for communication between communication nodes is shown in, a control frame (including a token frame) for transmitting a command, and a frame for transmitting data. In these frames, F is a flag (8 bits of 01111110), C is a command, DA is a communication destination address.
SA is a source address, FCS is a frame check sequence for error detection, and I is an information field in which data to be transmitted is stored. The token frame is
This is displayed because the C (command) of the control frame is a specific pattern (for example, all 1).

The communication node shown in FIG. 8 (70 to 70 shown in FIG.
In the configuration of 75), 80 is a bus interface unit,
Reference numeral 81 is a ring bus interface receiving unit that performs processing for receiving a signal from an upstream communication node, and 82 is identification of whether the frame is a token or a normal frame, control signals included in the frame, that is, an address and a command. A frame analysis unit that analyzes the data, such as a frame analysis unit 83 that receives from an upstream communication node and needs to be sent to another communication node, or a transmission control unit 86 that controls switching of the transmission data switching unit 86 when transmitting a frame from its own communication node. The control unit 84 switches between bypassing the frame received from upstream when the communication node has a failure and transmitting it to the downstream communication node, or transmitting the frame from the ring bus interface transmission unit 85 when there is no failure. , 85 is a ring bus interface transmission unit on the transmission side that performs processing for transmission to a downstream communication node, 6 is a transmission data switching unit for selecting either the frame from the frame analysis unit 82 or the frame generated from its own communication node, 87 is a processor interface unit for interfacing with the bus interface unit 80 and the processor side, and 88 is A processor unit of the communication node, and 89 is a memory unit of the communication node.

The bus interface unit 80 is provided in each communication node corresponding to the ring bus, and the processor unit 88 and the memory unit 89 connected via the processor interface unit 87 constitute a communication node. The operation of the communication node of FIG. 8 will be described below.

Normally, a frame from the ring bus is received by the ring bus interface receiver 81, and the frame analyzer 82 analyzes the received frame. When a certain communication node performs a transmission operation, the processor unit 88 issues a transmission request to the transmission control unit 83 via the processor interface unit 87. When the token frame detection signal is received from the frame analysis unit 82 in this state, the transmission control unit 83 causes the transmission data switching unit 86 to change to the processor unit 88.
And the data is transmitted from the ring bus interface transmitter 85 onto the ring bus, and after the final data transmission, the captured token frame is transmitted onto the ring bus to complete the transmission operation.

In such a ring bus system, there are one-to-one communication and one-to-n communication as the communication types between the communication nodes, which are B.1 of FIG. D of the frame structure shown in
It is identified by setting the communication destination address in the A section. As the communication destination address, an individual address is set for one-to-one communication, and a representative address, a group global address, and an all-global address are set for one-to-n communication. When the transmitted frame is received at the destination (or multiple destinations in the case of the representative address), the frame is picked up by the corresponding node, and the ACK frame indicating that it was received on the ring bus is sent to the source communication node. To do. When the transmission source receives this, it can be confirmed whether or not the frame is normally transmitted to the communication destination. Also, the destination communication node is
When reception is impossible due to buffer busy or the like, the REJ (rejection) frame is returned to complete the communication. The DA part of the token frame is set as an all-global address that any communication node can use.

Further, the communication management node 76 monitors the circulation time of the token circulating on the ring bus, and when the token cannot be detected within the specified time, notifies the host device as a token loss and reproduces a new token and rings. It has the function of sending data on the bus.

[0012]

As described above, in the ring bus system, the token frame that circulates on the ring bus must be captured when communication is performed. It will greatly affect your ability. Especially, during high traffic, each communication node has a transmission request, and if each communication node captures a circulating token frame and transmits data, the token hold time in each node increases, so the token However, there is a problem that the cycle time is extended and the processing capacity of the system is reduced.

According to the present invention, in a ring bus system in which a plurality of communication nodes are connected to a ring-shaped bus, it is possible to prevent a decrease in the processing capacity of the system due to an extremely delayed circulation time of token frames when the ring bus is congested. An object of the present invention is to provide a communication control method in a ring bus system that can obtain stable processing capacity without depending on traffic and ensure high reliability.

[0014]

FIG. 1 is a block diagram showing the principle of the present invention. In FIG. 1, 1 is a communication management node, 10 is a frame analysis unit that analyzes a received frame to detect a token frame (simply called a token), 11 is a normal token generation monitoring unit that generates and monitors a normal token,
Reference numeral 12 is a load state monitoring control means for monitoring the load state according to the circulation time of the normal token and controlling the generation / disappearance of individual tokens according to the degree of congestion when congestion occurs, 13-1 to 13-1.
13-n (n ≧ 1) are load state monitoring control means 1 respectively.
An individual token generation / monitoring unit that generates and monitors an individual token that can be used by a predetermined specific group of communication nodes (eg, a group with a large amount of communication via the ring bus) controlled by 2. Can be provided. 2 is a communication node, 20 is a frame analysis unit, 2
Reference numeral 1 is a normal token detecting means, and 22 is an individual token detecting means of its own group.

According to the present invention, when the circulation time of the normal token circulating on the ring bus is monitored to detect the congestion state of the bus by monitoring the load state, the individual token that can be used by a specific communication node group separately from the normal token. By generating one or a plurality of tokens according to the degree of congestion and sending them to the ring bus, the processing capability of the ring bus is improved.

[0016]

In the communication management node 1, a normal token is generated from the normal token generation / monitoring means 11 at the same time when the system is started, and the monitoring of the token circulation time is started. The frame analysis unit 10 analyzes a frame input from the ring bus, and when detecting a normal token, supplies it to the normal token generation monitoring unit 11. The normal token generation monitoring means 11 detects the patrol time from the previous transmission of the normal token to the ring bus to the reception each time, supplies it to the load state monitoring control means 12, and transmits the normal token to the ring bus.
When the detected patrol time continues to exceed the preset value, the load state monitoring control unit 12 determines that the congestion state exists, and drives the individual token generation monitoring unit 13-1 at the first stage.

The individual token generation monitoring means 13-1 generates an individual token of a type different from the normal token on the ring bus and sends it to the ring bus for monitoring. The individual token is given a priority (being available only to a specific group of communication nodes) according to a setting value that uses the DA unit (communication destination address) in normal communication in conformity with the frame structure of the normal token. For example, the DA part of the individual token can be set so that a group of communication nodes having a relatively large amount of communication on the ring bus can be used.

In the frame analysis unit 20 of the communication node 2, the normal token detecting means 21 detects the normal token, and when the normal token is detected when there is a transmission request from this communication node, the transmission right is obtained and the data is transmitted. To do. Further, the individual token detecting means 22 detects whether or not the individual token is assigned to the communication node 2 as an available token. Specifically, the communication node (group) is previously stored in the individual token detecting means 22.
The address of the individual token assigned to is set, and when the set individual token is detected, this communication node 2 obtains the transmission right and can transmit the data, and when it finishes, the individual token is transmitted. .

When the load state monitoring control means 12 detects that the congestion state continues to exceed the set value corresponding to the bad state even after the individual token generation monitoring means 13-1 is activated due to the occurrence of the congestion state. , And activates the second individual token generation monitoring means 13-2. The individual token generation / monitoring unit 13-2 also has the same configuration as the individual token generation / monitoring unit 13-1 that operates first, and generates and monitors an individual token that can be used by another communication node group. In this way, the plurality of individual token generation monitoring means 13
Individual tokens are generated by -1, 13-2, ... 13-n (not shown).

When the congestion state is sequentially resolved by monitoring the circulation time of the normal tokens, the load state monitoring control means 12 operates in response to the individual token generation monitoring means 13-n ... 13-2, 13-1 is stopped in order to extinguish each individual token.

[0021]

DETAILED DESCRIPTION FIG. 2 is a block diagram of the essential parts of a communication management node, and FIG.
FIG. 3 is a diagram showing a specific example of each token frame.

In this embodiment, there are two individual tokens, an individual token A that occurs in a lightly congested state and an individual token B that occurs when the congestion state worsens even if the individual token A occurs. I shall.

FIG. 2 mainly shows a portion related to token management of the communication management node. In the figure, 30 is a frame analysis unit, 30a is a DA (communication destination address) unit analysis unit, and 30b.
Is a C (command) section analysis section 30b, 31 is a token management section for managing a normal token and an individual token according to the present invention, 32 is a normal token generation / monitoring section, 32a is a monitoring counter, 33 is a load state monitoring control section, 33a Is a congestion counter A, 33b is a congestion counter B, 33c is a congestion state A
Is a flag indicating whether it is on or off, 33d is a flag indicating whether the congestion state B is on or off, and 34 is an individual token A generation /
A monitoring unit, 34a is a monitoring counter that monitors the circulation time of the individual token A and detects an abnormality, 35 is an individual token B generation / monitoring unit, 35a is a monitoring counter that monitors the individual token B, 36 is a transmission control unit, and 37 is It is a transmission data switching unit.

The communication management node is basically constructed by the same mechanism as a general communication node, and the frame analysis unit 30, the transmission control unit 36 and the transmission data switching unit 37 of FIG. 2 are the same as those of FIG. Corresponding to the names 82, 83, 86, other parts of the communication node (81, 84, 8 in FIG. 8)
5, 87-89) are omitted in FIG.

A specific example of the token frame is shown in FIG.
The basic structure of the normal token frame and the individual token frame according to the present invention is the same as that of the conventional control frame (see B. in FIG. 7), F (flag), C (command), DA.
(Destination address), SA (source address), FCS
(Frame check sequence) and F (flag).

Token frame structure C (command)
The section indicates that it is a token (for example, “FF (H)”, where H is hexadecimal display), and the DA section indicates whether it is a normal token or an individual token frame. In this example, as shown in FIG.
Part A is "FF (H)" and individual token A is "E *"
(H), that is, when displayed in 8 bits, "1110 **
** ”(however, * is ignored), and when individual token B is displayed as“ C * (H) ”, that is, 8 bits,“ 11 ”
00 *** ”. As described above, when the normal token and the individual token are set, when the DA part is analyzed in order to identify the type of the token, it can be identified by only the first 4 bits.

FIG. 4 is a control flow in the configuration shown in FIG. 2, FIG. 5 is a diagram showing an operation example of a normal token and an individual token on the ring bus, and FIG. 6 is a diagram showing a change in circulation time of the normal token. It is explanatory drawing of a congestion state.

The control flow of FIG. 4 is changed to that of FIG. 2, FIG. 5 and FIG.
Will be described with reference to. In the operation example of FIG. 5, the communication nodes # 1 to # 6 (however, # 4 to # 6 shown in FIG. 7 are included.
Shows the operation of each token in the ring bus system consisting of (1) normal state, (2) congestion state A, (3) congestion state B in the ring bus system consisting of the communication management node # 7. Reference numeral 6 shows the relationship between the traveling time and the specified time corresponding to the states (1) to (3) in FIG. Further, in this embodiment, each communication node # 1 to # 6 is divided into processing unit groups of the system, and only the group of communication nodes # 4 to # 6 can use the individual token frame A and the communication nodes # 1 to # are processed. Only the group of 3 can use the individual token B, and the priority is added to each individual token.

When a frame is first received by the communication management node (S1 in FIG. 4), the frame analysis unit (30 in FIG. 2).
The DA unit (communication destination address) is analyzed by the DA unit analysis unit (30a in FIG. 2) (S2). Regarding this result, it is determined whether the received frame is addressed to the own node (at step S3). in this case,
The device number is assigned to each communication node in advance, and it is determined whether or not the node is addressed to its own node by matching the value with the value specified by the DA unit. However, the DA part of the normal token is "F *", and any communication node is regarded as a match. Further, the individual tokens A and B are designated as "E *" and "C *", and it is considered that the grouped nodes of the respective communication nodes match.

If it is not addressed to its own node, the frame is
The data is sent through to the ring bus through, and then sent to the next connected communication node (at step S4). When it is addressed to its own node, the C part is analyzed (at step S5), and whether it is a normal token frame is identified by the analysis result of the part C and DA part (at step S6).

In this explanation, first, (1) in FIGS.
It is assumed that the normal token is present and only the normal token exists on the ring bus. Here, when a control frame (a frame instructing a command) for the communication management node is received instead of the normal token frame,
S7 (individual token A), S10 (individual token B)
Each of the above judgments is NO, and the processing corresponding to the command is executed (at step S13).

In the case of a normal token frame, a monitoring counter of the normal token generation / monitoring unit 32 (32 in FIG. 2).
Although a) is read, its value is analyzed (S14 in FIG. 4), the value is reset, and at the same time, although not shown in FIG. Monitoring counter (Fig. 2
32a) always counts the patrol time from generation of the normal token to patrol and return of the ring bus, and determines whether the analyzed time value of the time value of the monitoring counter is within the specified value A (S15 in the same figure). ). This specified value A is shown as the specified time A in FIG. 6, and the patrol time is normally within the specified value A in normal times. If it is judged to be within the specified value, the congestion counter A
(33a in FIG. 2) is reset (S16 in FIG. 4), and the congestion state A (33c in FIG. 2 is ON, the individual token A is generated.
If the state in which the monitoring unit 34 is driven to generate the individual token A is ON, it is turned OFF (S
17), preparing for the next frame reception.

If the value of the monitoring counter is not within the specified value A, then it is judged whether it is within the specified value B (S18 in FIG. 4). This specified value B is also shown as the specified time B in FIG. It's a longer time than. If it is within the specified value B in this determination, the congestion counter B (33b in FIG. 2) is reset (S19 in FIG. 4), the congestion state B (33d in FIG. 2 is on, and the individual token B generation / monitoring unit 35 When driven to generate the individual token B) is turned on, it is turned off (at step S20) and the congestion counter A is incremented by +1 (at step S2).
1). Further, it is determined whether the congestion counter A overflows (exceeds a preset value) (S22), and if it overflows, an individual token A is generated and sent to the ring bus, and the congestion state A (see FIG. 2). The flag of 33c) is turned on (at steps S23 to S25) and the next reception frame is waited for.

As described above, the individual token A has the specified time A
It is generated when the number of times exceeds a certain number of times continues. This state is shown as (2) congestion state A in FIGS. 5 and 6. In this example, the generated individual token A is as shown in FIG.
As shown in (2) of the above, priority is added to the group of the communication nodes # 4 to # 6 (the group of communication nodes having a lot of data transfer traffic), and the other communication nodes # 1
~ # 3 is not available.

Next, if the monitoring counter value is not within the specified value B in S18, the congestion counter B is incremented by 1 (S26 in the same step), and it is determined whether the value overflows (S27 in the same step). The token B is generated and sent to the ring bus, and the congestion state B (33 in FIG.
The flag of d) is turned on (at steps S28 to S30).

Also in this case, the individual token B has the specified time B.
It is generated when the number of times exceeds a certain number of times continues. This is shown as (3) congestion state B in FIGS. 5 and 6. In this example, the generated individual token B is (3) in FIG.
As shown in, the priority is added to the communication nodes # 1 to # 3, and other communication nodes cannot be used. In this state (3), the normal token, the individual token A, and the individual token B exist on the ring bus.

Next, the individual token A or the individual token A
If both of the individual token B and the individual token B exist on the ring bus, it is determined in S6 of FIG. 4 that the frame is not a normal token frame, and then it is determined whether the individual token A (S7 of FIG. 4). If there is, it is determined whether or not the congestion state A flag (33c in FIG. 2) is on (at step S8).
If it is on, the individual token A is sent to the ring bus as it is (not shown), and if it is off by the operation of S17 described above, that is, if the load is reduced from the congestion state A to the normal state, the individual token A is transmitted. The token A is erased (at step S9).

The received token is the individual token A.
If not, it is determined whether it is the individual token B (at step S10), and if it is the individual token B, the flag of the congestion state B (33 in FIG. 2).
If d) is on (S11), if it is on, the individual token B remains as it is (not shown), and if it is off (congestion state B to congestion state A or normal state, the load is reduced). Case), the individual token B is erased (FIG. 4).
S12).

Each communication node receives a normal token frame and one or a plurality of individual tokens, but each communication node is divided into groups in advance in the design of the ring bus system, and each communication node determines which group. Is set or is set, the individual token that can be used in the own node can be identified in the frame analysis unit when a frame is received in the individual token reception process. In the examples shown in (2) and (3) of FIG. 5, the individual token A is through in the communication nodes # 1 to # 3, but can be used in the communication nodes # 4 to # 6. When a usable individual token is detected, it is processed in the same way as a normal token, and if there is a transmission request at that time, the individual token can be held and sent, and the individual token held after sending To the ring bus.

[0040]

According to the present invention, in a ring bus system for performing communication by obtaining a transmission right by capturing a token frame circulating on the ring bus, a token frame in each communication node according to an increase in traffic Even if the hold time of the token becomes long, the load condition on the ring bus is detected early and the individual token frame is sent out on the ring bus according to the load condition, so that the token frame can be captured at each communication node. Since it can be increased, stable processing capacity independent of traffic can be obtained. This makes it possible to realize a highly reliable ring bus system.

[Brief description of drawings]

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

FIG. 2 is a configuration diagram of a main part of a communication management node.

FIG. 3 is a diagram showing a specific example of each token frame.

FIG. 4 is a diagram showing a control flow in the configuration shown in FIG.

FIG. 5 is a diagram showing an operation example of a normal token and an individual token on a ring bus.

FIG. 6 is an explanatory diagram of a congestion state with respect to a change in circulation time of a normal token.

FIG. 7 is an explanatory diagram of a ring bus system.

FIG. 8 is a configuration example of a communication node.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 communication management node 10 frame analysis unit 11 normal token generation monitoring unit 12 load state monitoring control unit 13-1 to 13-2 individual token generation monitoring unit 2 communication node 20 frame analysis unit 21 normal token detection unit 22 individual token detection unit

Claims (4)

[Claims] 1. A ring bus system for connecting a plurality of communication nodes with a ring bus to obtain a transmission right by a token passing system, which comprises a communication management node for managing at least one system, wherein the communication management node has a ring bus. A normal token generation / monitoring unit that generates one normal token frame that circulates above and monitors the circulation time on the ring bus, and inputs the circulation time of the normal token frame to continue the preset time. When the ring bus exceeds the congestion state, the load state monitoring control means determines that the congestion state occurs, and the congestion state determination output generates an individual token frame with priority added separately from the normal token frame and transmits it to the ring bus. A ring bus system, characterized in that it comprises an individual token generation monitoring means for monitoring. Communication control method. 2. The claim 1, wherein a plurality of the individual token generation monitoring means are provided, and the load status monitoring control means is set in advance with a prescribed time of a plurality of patrol times corresponding to a plurality of stepwise congestion states. , The regular token frame on the ring bus continues, and it is sequentially determined whether or not the specified time value of each set stage is continuously exceeded, and if it exceeds, a corresponding congestion state is set, and A communication control method in a ring bus system, wherein a plurality of individual token generation / monitoring means are driven according to the setting of the congestion state at each stage, and generate and monitor corresponding individual token frames. 3. The load state monitoring means according to claim 1 or 2, further comprising a congestion counter corresponding to each congestion state, and a congestion counter corresponding to a congestion time exceeding a prescribed time in comparison with the prescribed time of each congestion state. Is added to set a corresponding congestion state when the count value exceeds a preset value, and when the comparison time does not exceed the specified time, the corresponding congestion counter is reset and the congestion status corresponding to the specified time is A communication control method in a ring bus system characterized by canceling the setting when set. 4. The communication node according to claim 1 or 2, wherein each communication node includes a frame analysis unit that analyzes a received frame from a ring bus, and the frame analysis unit includes a normal token detection unit that detects the normal token, and An individual token detecting unit for detecting an individual token available to the communication node, and each communication node performs transmission using the normal token or the individual token detected by the normal token detecting unit or the individual token detecting unit. Communication control method in the characteristic ring bus system.