JP3062904B2 - Data transmission method for distributed computer system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data transmission system of a distributed computer system, and more particularly to a data transmission system of a distributed computer system for periodically transmitting periodic information from a lower computer to an upper computer.

[0002]

2. Description of the Related Art Conventionally, when transmitting periodic data from a lower-level computer to a higher-level computer, all the lower-level computers independently transmitted directly to the upper-level computer. Therefore, in receiving the one cycle data of each lower computer, the upper computer side needs the same number of data reception processes as the number of lower computers.

[0003]

In the above-mentioned prior art, with respect to the periodic data transmitted from the lower computer to the upper computer, all the lower computers transmit the data to the upper computer. That is, the load of the transmission / reception processing of the upper computer increases in proportion to the number of lower computers. Here, assuming that the time required for the host computer to perform one data reception process from the network is a and the time required to decode one piece of information of the received message is b, n pieces of information are received in one transmission data reception. When included, the time required for the transmission data reception processing is a + nb
In addition, the time required for reception processing when receiving a message including n pieces of information m times in one transmission data reception can be represented by ma + mnb. In this way, it is impossible to reduce the number of data that needs to be processed. Therefore, it is effective to reduce the number m of transmissions to the host computer as much as possible to reduce the load on the host computer. Recognize. The effect is great when a is large. An object of the present invention is to reduce the number of transmissions of periodic data transmitted from a lower computer to a higher computer, reduce the load on the higher computer, and improve the processing capacity other than the transmission processing.

[0004]

In order to achieve the above object, a group of lower computers is divided into several groups, and each lower computer in the group is uniquely assigned a level. Instead of transmitting the periodic data directly to the upper computer, the periodic data is transmitted to the upper-level lower computers in the group, and only the highest-level lower computer in the group is configured to transmit the periodic data collectively to the upper computer. . When transmitting periodic data to a higher-level computer, a storage area for periodic data is provided in the lower-level computer to prevent the collection time of the periodic data from being combined at each level of the lower-level computer. Is separated. After the cycle data is stored in the storage area in the collection process, a time difference is set in the transmission process between the timings at which the cycle data of each lower computer is transmitted, and the collection time of the cycle data combined in each group is set for each lower computer. Is configured to transmit the periodic data to the host computer so as to be equal to

[0005]

Operation The above operation will be described with reference to FIG. n-
One lower-level computer 100 is divided into k groups of three, and a level is uniquely added to each lower-level computer 100 in each group. Among them, the lower-level computers 100 of the group 1 are assigned node NOs = 2, 3, 4 in ascending order of level.
And Also, the host computer 300 is set to node NO = 1. At this time, the flow of the periodic data of the lower computer 100 is such that the node NO = 2 which is the lowest level is transmitted to the node NO = 3 which is the upper level, and the node NO which has received it is transmitted.
In the case of = 3, it transmits to the node NO = 4 which is the upper level together with the cycle information of the node NO = 2. Then, at node NO = 4 receiving this, it is determined that the self node is at the highest level, the cycle information of nodes NO = 2 and 3 is combined with its own cycle information, and the node which is the upper-level computer 300 Send to NO = 1. According to this transmission method, the transmission of the periodic data to the host computer in the number of nodes, that is, n-1 times, is now completed in the number of groups, that is, k times. By the way, when all the periodic data transmissions from each node need to be performed at the same time, the latest periodic information at each node is not immediately transmitted to the upper computer, and the time of the periodic data combined at each level of the lower computer is Each transmission may be shifted. Therefore, in order to prevent the delay of the cycle data from the lower computer to the higher computer, the cycle information of each lower computer is temporarily stored in the storage area, respectively.
After that, a time difference is provided for the timing of the periodic data transmission time for each node of each level in each group so that data can be communicated between the nodes, and the transmission time timing of the periodic data is made earlier for lower-level nodes. In addition, there is no lag in the collection time of the periodic data combined at each level of the lower computer.

[0006]

Embodiments of the present invention will be described below. FIG.
1 is an embodiment of the present invention, showing an overall system configuration. HDLC transmission line 2 from each substation 400 dispersed in a wide area
01, a subordinate computer 100 for transmission control, which shares some functions of the host computer 300 for control and independently executes processing of the connected substations 400; Host computer 30 for centralized monitoring and control of
0 is connected and configured by a network 200 for transmitting and receiving information. The host computer 300 includes a computer main body 301, a CRT display device 302, and a keyboard 30.
3 Information from each substation 400 is sequentially transmitted to the lower-level computer 100 via the transmission line 201, edited by the lower-level computer 100, and broadcast to the network 200 if necessary. The host computer 300 that has received it processes and processes it on the computer main body 301 and displays it on the CRT display device 302. That is, the host computer 300 needs to grasp the ever-changing operation information of each substation 400 in real time and promptly and accurately display and report it to the operator. Therefore, in each substation 400, the latest data is The information is periodically transmitted to the host computer 300, and the shorter the transmission cycle, the greater the novelty of the information. Information exchange between the lower-level computers 100 is performed via the network 200. Here, the measurement information of each of the substations 400 is transmitted on a transmission line 20 every 1 second.
An example will be described in which the lower computer 100 that is transmitted to and received by the lower computer 100 via the network interface 1 transmits the information to the network 200 in the same cycle, and the higher computer 300 that receives the information processes the information and then displays the information on the CRT 302. .

[0007] As shown in FIG.
On the other hand, when N−1 lower computers 100 are connected by the network 200, the upper computer 300 is connected to the node NO = 1, and the lower computers are connected to the node NO =
.., N. When N-1 units are divided into three units, each becomes a group number = 1, 2,..., K. However, since the number of one group does not need to be equal in all groups, when N-1 is not divisible by 3, a group having a number different from that of the other groups is generated. Then, as shown in FIG. 2, the flow of the cycle information is as follows: node NO = 2 → node NO = 3 → node NO = 4 → node NO = 1, node NO = 5 → node NO = 6 → node NO = 7 → Node NO = 1, and generally, node NO = 3n−1 → node NO = 3n → node NO = 3n
+ 1 → node NO = 1 (where n = 1, 2,..., (N−
1) / 3).

[0007] Group 1 (node NO =
2, 3, 4), the transmission / reception operation will be described with reference to FIGS. 4, 5, 6, and 7. FIG. First, a node state management table 109 for managing the state of each node is provided as shown in FIG. For node NO = 2, 3, 4, the node level is set to 1, 2, 3, and the periodic data transmission time is set to 0.90 seconds, 0.95 seconds, 1.00, respectively.
Defined as seconds. In addition, all the operation / stop states of the nodes are set to operation (RUN).

FIG. 4 is a functional block diagram of the lower-level computer 100. In FIG. 4, cycle information from the substation 400 is transmitted to the lower-level computer 100 via the transmission line 201. In the lower computer 100, the information is received by the transmission path receiver 112 and stored in the reception data area (lower) 113. In response, the receiving and editing process (lower order) 114
Transfers the contents of the reception data area (lower) 113 to the reception data buffer (lower) 115 and activates the upstream message analysis processing 116. Then, in the upstream message analysis processing 116, the data is determined, and the data in the reception data buffer (lower) 115 is transferred to the upstream data reception / transmission data buffer (higher) 117. Here, for the node NO = 2, the periodic transfer process 108 started at a one-second cycle sends a transmission request of the periodic information to the transmission editing process 106 at 0.90 seconds per hour. In the transmission editing process 106, since the start request is received from the periodic transfer process 108 and the period information does not exist in the downlink data reception transmission data buffer (upper) 107, only the period information of the uplink data reception transmission data buffer (upper) 117 is stored. Since the self-node is not at the highest level according to the node status management table 109, the higher-level node NO = 3 is set as the transmission destination, and is written into the transmission buffer 110 in the format shown in FIG. In the format of FIG. 7, the transmission source is node NO = 2, the transmission destination is node NO = 3, the request content is transmitted,
The data type is periodic, the serial number is 1 (described later), and the node NO =
2 measurement information. The network transmission unit 111
The serial number 1 for network transmission / reception management is added to the data in the transmission buffer 110 and the data is transmitted to the network 200 by broadcasting. Next, at the node NO = 3, the data of the serial number 1 on the network 200 is received by the network receiving unit 101, and the transmission destination is determined to be the data addressed to the own node at the node NO = 3.
2 and serial number 1 in the format of FIG.
Is normally received, the transmission source node N
O = 2 is transmitted to the network 200 as a transmission destination. In the format of FIG. 7, the transmission destination, the transmission source, and the serial number need to match those of FIG. 7, and ACK indicates a reply. In addition, the reception editing process (upper) 103 moves the contents of the reception data area (upper) 102 to the reception data buffer (upper) 104 and activates a downstream message analysis process 105. Then, in the downlink message analysis processing 105, the data is determined, and the data in the reception data buffer (upper) 104 is transferred to the downlink data reception / transmission data buffer (upper) 107. For the node NO = 3, the periodic data sent from the substation 400 to the lower-level computer 100 via the transmission line 201 is the same as the reception data from the transmission line of the node NO = 2. The data is transferred to the buffer (upper) 117. The periodic transfer processing 108 issues a transmission request of the periodic information to the transmission editing processing 106 at 0.95 seconds per hour, such as after receiving the periodic data from the node NO = 2. In the transmission editing process 106, upon receiving a start request from the periodic transfer process 108, the cycle information of the downlink data reception / transmission data buffer (upper) 107 and the cycle information of the uplink data reception / transmission data buffer (upper) 117 are added to the node. Since the self-node is not at the highest level according to the state management table 109, the node 4 at the higher level is set as the transmission destination and is written into the transmission buffer 110 in the format shown in FIG. The network transmission unit 111 adds the serial number 2 to the data in the transmission buffer 110 and transmits the data to the network 200 by broadcasting. Then, at the node NO = 4, the data of the serial number 2 on the network 200 is transmitted to the network receiving unit 10.
1 and the transmission destination is determined to be the data addressed to the own node when the node number is NO = 4, stored in the reception data area (upper) 102, and that the data of serial number 2 is normally received in the format of FIG. Is transmitted to the network 200 with the transmission source node 3 as the transmission destination. In addition, similarly to the process of receiving the cycle information from the network 200 of the node NO = 3, this data is transferred to the downstream data reception / transmission data buffer (upper) 107. Further, node NO
= 4, the periodic data sent from the substation 400 to the lower-level computer 100 via the transmission line 201 also receives the upstream data reception transmission data buffer (upper) as in the reception process from the transmission line of node NO = 2. 117.
The periodic transfer processing 108 issues a transmission request of the periodic information to the transmission editing processing 106 at 1.00 seconds every hour, such as after receiving the periodic data from the node NO = 3. In the transmission editing process 106, upon receiving a start request from the periodic transfer process 108, the cycle information of the downlink data reception / transmission data buffer (upper) 107 and the cycle information of the uplink data reception / transmission data buffer (upper) 117 are added to the node. It is determined from the status management table 109 that the self node is at the highest level, and the format is as shown in FIG.
Write to the transmission buffer 110. The network transmission unit 111 adds the serial number 3 to the data in the transmission buffer 110 and transmits the data to the network 200 by broadcasting. Then, node NO = 1, which is the host computer 300
Receives the data of serial number 3 on the network 200,
When the transmission destination is the node NO = 1, it is determined that the data is addressed to the own node, and the data is acquired.
Is normally received, the transmission source node N
O = 4 is transmitted to the network 200 as a transmission destination. Then, after processing and processing by the computer main body 301, C
The measurement data of the period information is updated on the RT display device 302.

By the way, if all the transmission timings of the period information of each node are set to the same time of the standard clock,
, Are transmitted at the same time, the data of the node NO = 4 is immediately transmitted to the node NO = 1, but the data of the node NO = 3 is transmitted about one second later, and the data of the node NO =
Since the data of No. 2 is transmitted to the node NO = 1 after about 2 seconds, the credibility of the periodic data may be impaired. Therefore, the node state management table 109 shown in FIG.
A time difference is provided in the periodic data transmission time of each node, and after receiving the periodic data from the lower-level node, the data is transmitted to the next node together with its own data, thereby preventing the delay of the periodic information.

Next, an operation when the node 3 is stopped will be described with reference to FIGS. The operating status of each node is
As shown in FIGS. 8 and 9, when the node state management table has the same configuration as in FIGS. 5 and 6, first, at the node NO = 2, the lower-level computer is connected from the substation 400 via the transmission line 201. The periodic data sent to 100 is transferred to the upstream data receiving / transmitting data buffer (upper) 117. The periodic transfer process 108, which is activated in a one-second cycle, issues a transmission request for period information to the transmission editing process 106 at 0.90 seconds per hour. In the transmission editing process 106, the start request is received from the periodic transfer process 108, and since there is no cycle information in the downlink data reception transmission data buffer (upper) 107, the uplink data reception transmission data buffer (upper) 117.
Since the node itself is not at the highest level in the node state management table 109, a search is made for a higher-level node, and node NO = 3 is stopped. Write to buffer 110. The network transmission unit 111 stores the serial number 1 for network transmission / reception management in the data in the transmission buffer 110.
, And broadcast the broadcast to the network 200. Then, at node NO = 4, network 2
00 is received by the network receiving unit 101, the destination is determined to be the data addressed to the own node when the node number is NO = 4, and stored in the reception data area (upper) 102, and the data of the serial number 1 is stored. The normal reception is transmitted to the network 200 with the transmission source node NO = 2 as the transmission destination. This reception data area (upper) 10
2 is transferred to the downstream data receiving and transmitting data buffer (upper) 107. At node NO = 4, the periodic data sent from the substation 400 to the lower computer 100 via the transmission line 201 is transferred to the upstream data receiving / transmitting data buffer (upper) 117. The periodic transfer processing 108 issues a transmission request of the periodic information to the transmission editing processing 106 at 1.00 second every hour, such as after receiving the periodic data from the nodes NO = 2 and 3.
In the transmission editing process 106, upon receiving a start request from the periodic transfer process 108, the cycle information of the downlink data reception / transmission data buffer (upper) 107 and the cycle information of the uplink data reception / transmission data buffer (upper) 117 are added to the node. The status management table 109 determines that the own node is at the highest level, and writes the result to the transmission buffer 110. The network transmission unit 111 includes the transmission buffer 11
The serial number 2 is added to the data of 0, and the data is transmitted to the network 200 by broadcasting. And the host computer 300
The node No. 1 receives the data of the serial number 2 on the network 200, determines that the transmission destination is the data addressed to the own node with the node NO = 1, obtains the data, and normally receives the data of the serial number 2 From the transmission source node NO =
4 to the network 200 as the destination. Then, after processing and processing are performed by the computer main body 301, the measurement data of the cycle information is displayed and updated on the CRT display device 302.

FIG. 10 and FIG. 11 show the processing flow of these procedures. In FIG. 10, in the lower-level computer 100,
First, the periodic data from the network 200 is received and edited (step 10). Also, the periodic data transmitted from the substation 400 via the transmission line 201 is received and edited (step 11). Subsequently, the downstream message from the network 200 is analyzed (step 12). In addition, the upstream message is analyzed (step 13).
The downlink message and the uplink message are cyclically transferred (step 14), and transmission data to be transmitted to the network 200 is edited (step 15). Next, FIG.
1, a downlink message from the network 200 is acquired if it is data addressed to the own node from the lower-level lower-level computer 100.
As for the upstream message from No. 1, it is obtained as it is, and the respective data are combined (step 20). Among them, first, the own node and the group number are obtained (step 21). Next, it is determined whether or not the own node is the highest-level node in the group (step 22). If the node is at the highest level, the host computer (node 1)
Is regarded as the transmission destination of the periodic data (step 23).
If the node is not the highest level node, it is determined whether the higher level node is operating in the group (step 24). If the upper-level node is not operating, the higher-level computer (node 1) is regarded as the transmission destination of the periodic data (step 25). If the upper-level node is not operating, the periodic data is transmitted to the upper-level lower-level computer (step 25). By performing the above, one of the cycle information received by the host computer 300
Although the number of messages per transmission increases by three times, the number of receptions by the computer is reduced to one third. By the way, in general, the load of transmission and reception between computers is affected by the number of transmissions and receptions rather than the amount of messages unless the amount of messages is enormous, and the load increases as the number of nodes connected to the network increases. Therefore, according to the present embodiment, the load on the host computer 300 can be greatly reduced, and the processing capacity other than the transmission processing can be significantly improved. Even if the number of nodes connected to the network increases, the load on the host computer 300 can be suppressed to a minimum increase by appropriately setting the number of groups. Further, in the present embodiment, for information requiring urgency such as accident data from the substation 400, the host computer 3
00 can be easily transmitted directly.

[0012]

According to the present invention, the number of periodic data transmissions from the lower computer to the upper computer can be reduced, thereby reducing the load on the upper computer and improving the other processing capacity accordingly. . Further, according to the present invention, even if a lower computer is added due to the addition of a substation, the number of transmissions of periodic data does not increase, so that the load on the upper computer can be suppressed to a minimum increase.

[Brief description of the drawings]

FIG. 1 is a configuration of a distributed system according to an embodiment.

FIG. 2 shows a flow of periodic data between systems according to an embodiment of the present invention.

FIG. 3 is a diagram illustrating a group configuration of a distributed system according to an embodiment;

FIG. 4 is a functional block diagram of a lower computer main body;

FIG. 5 shows an example of a node status management table in a lower computer.

FIG. 6 shows a flow of periodic data for one group in the embodiment.

FIG. 7 is a format of periodic data.

FIG. 8 is an example of a node status management table in a lower computer when a node stops;

FIG. 9 illustrates a flow of periodic data for one group according to the embodiment when the node is stopped.

FIG. 10 is a flowchart of periodic data processing of a lower computer.

FIG. 11 is a flowchart of each node state determination process in the flowchart.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 Lower computer 200 Network 201 HDLC transmission line 300 Upper computer 301 Upper computer main body 302 CRT display device 303 Keyboard 400 Substation 101 Network receiver 102 Receive data area (upper) 103 Reception edit processing (upper) 104 Receive data buffer (upper) 105 Downlink message analysis 106 Transmission editing processing (upper) 107 Downlink data reception / transmission data buffer (upper) 108 Periodic transfer processing 109 Node status management table 110 Transmission buffer (upper) 111 Network transmission unit 112 Transmission path reception unit 113 Received data area (Lower) 114 Reception edit processing (Lower) 115 Received data buffer (Lower) 116 Upstream message analysis 117 Upstream data reception / transmission data buffer (upper) 118 Transmission data Tabaffa (lower) 119 sends editing process (lower) 120 transmission buffer (lower) 121 transmission line transmission unit

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-3-211646 (JP, A) JP-A-2-259848 (JP, A) JP-A-2-95044 (JP, A) JP-A-63- 214038 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06F 13/00 G06F 12/00

Claims (4)

(57) [Claims] 1. A system comprising a plurality of independent lower computers, at least one upper computer, and a network capable of transmitting and receiving information between the plurality of computer systems. A distributed computer system for transmitting periodic information, wherein the periodic information of one lower computer is transmitted to another lower computer, and only the representative lower computer collectively transmits the periodic information to the upper computer. Data transmission method for computer systems. 2. The computer system according to claim 1, wherein the plurality of lower computers, each of which is independent of each other, are classified into any number of groups, and a lower computer in the group is uniquely assigned a level. A data transmission method for a distributed computer system, wherein periodic data is transmitted to a lower-level computer. 3. The low-level computer according to claim 1, further comprising a storage area for storing period information, storing the period information in the storage area during data collection processing,
A data transmission method for a distributed computer system characterized by performing a transfer process. 4. The method according to claim 1, wherein when transmitting the cycle information from the lower-level computer to the higher-level computer, each of the lower-level computers in the group has the same cycle information collection time. A data transmission method for a distributed computer system, characterized in that a time difference is provided in the timing of transmitting period information of a lower-level computer.