JPH0560699B2 - - Google Patents

Description

[Detailed description of the invention] [Industrial application field] This invention relates to an information transmission system.

[Conventional technology]

FIG. 4 is a block diagram showing the configuration of a conventional information transmission system. In the figure, A is an information processing device of station A consisting of a computer system;
B is an information processing device of the B station consisting of a computer system, and T is an information transmission device between both information processing devices A and B. The information transmission device T is composed of a communication channel a on the information processing device A side, a communication channel b on the information processing device B side, and a transmission path l connecting both channels a and b. M1 is a shared memory between information processing device A and communication channel a, and M2 is a shared memory between information processing device B and communication channel b. Each shared memory M1, M2 is a reception buffer.
It includes RB1, RB2 and transmission buffers TB1, TB2.

Next, the operation will be explained. When transmitting information from information processing device A to information processing device B, information processing device A uses transmission buffer TB1 of shared memory M1.
, and sends a transmission request signal S1 to communication channel a. Communication channel a transmits data to communication channel b via transmission path l. When the communication channel b receives the data, it writes it into the reception buffer RB2 of the shared memory M2, and outputs the data reception signal S2 to the information processing device B. After reading the data, information processing device B issues a reading completion signal S3 to communication channel b. Communication channel b issues a transmission/reception procedure completion signal to communication channel a. Communication channel a issues an information transmission completion signal S4 to information processing device A.

When transmitting in the opposite direction, do the same.

[Problem that the invention seeks to solve]

Conventional information transmission methods were performed as described above, so when information transmission is considerably slower than information processing, it is effective because information transmission becomes a bottleneck for the entire processing function, but information transmission is If the speed is sufficiently fast, the information processing device must perform interrupt processing for the data reception signal and information transmission completion signal from the information transmission device, and the above-mentioned communication between the information processing device and the information transmission device is frequently performed. When the handshake was performed, the arbor head of the software became large, which caused a problem of lowering the efficiency of the information processing as a whole.

This invention was made to solve the above-mentioned problems, and it eliminates the control signal between the information processing device and the information transmission device, and replaces it with the function of the software on the information processing device. The purpose is to obtain an information transmission method that can be obtained by

[Means for solving problems]

In the information transmission method according to the present invention, the information processing device on the transmitting side writes the transmission data to the transmission buffer on the shared memory with the information transmitting device at an arbitrary timing. The information transmission device does not perform transmission according to a request from the information processing device on the sending side, but performs this automatically and periodically. The receiving side information processing device writes the received data to a receiving buffer on a shared memory with the information transmitting device at an arbitrary timing. Data writing/writing from the perspective of the information processing device
In order to match the number of reads and the contents, a part of the transmitted data is used as synchronization data, and the receiving side sends this data back, and the information processing is performed by referring to this data by the software on each information processing device. Synchronize between devices.

[Effect]

Synchronization in the present invention is achieved by mutually referencing synchronization data, which is part of the contents of transmission data that is periodically transmitted, on the information processing apparatus side, and the progress status of the transmission procedure is known.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the system configuration of the information transmission system of the present invention, and shows the same or equivalent parts as FIG. 4 showing the prior art described above. The difference between the configuration of this embodiment and the configuration of the prior art described above is that there is no handshake control line.

Next, the operation of this embodiment will be explained with reference to the flowcharts shown in FIG. 1 and FIGS. 2A and 2B.

Communication channel a and communication channel b alternately use transmission buffers TB1 and TB in shared memories M1 and M2.
Send the contents of 2. Received data is received buffer
Enter RB1 and RB2. It is preferable that the transmission cycle be as short as the transmission capabilities of communication channels a and b allow.

Information processing device A sends data 1 to information processing device B
If you want to pass it twice, the procedure is as follows.

The information processing device A writes data to the transmission buffer TB1 of the shared memory M1, and adds "+1" to the synchronization data (=P1).

By the action of the information transmission device T, the contents of the transmission buffer TB1 of the shared memory M1 are copied to the reception buffer RB2 of the shared memory M2 of the information processing device B (P1=P2).

As shown in the reception processing flowchart of FIG.
By repeatedly referring to the contents of the synchronization data P2, a change in the value of the synchronization data P2 is detected, and when there is a change, it is read as new data has arrived, and then the value of the synchronization data P2 is sent to the buffer TB.
2 as P3 (P2→P3).

By the action of the information transmission device T, the contents of the transmission buffer TB2 of the shared memory M2 are copied to the reception buffer RB1 of the shared memory M1 of the information processing device A (P3=P4).

Information processing device A learns that information processing device B has taken the data because the synchronization data becomes P4=P1.

In other words, on the sending side, the value of the synchronization data at the time of sending and the value returned are equivalent, so that
Know the receiver's completion.

On the receiving side, notification is made by referring to the synchronization data in the received data and sending back new synchronization data.

When the above information transmission method is used in an N-to-N bus type network, since this method originally manages the use of transmission paths (they are used alternately automatically), it is relatively easy to connect N-to-N. Functions can be expanded to.

FIG. 3 shows the shared memory M1 to each information processing device of stations A to D when the information transmission method shown in FIG. 1 is expanded.
FIG. 3 is a block diagram showing the configuration of only M4. 1st
When expanding the one-to-one communication shown in the figure to N-to-N as shown in Fig. 3, each station uses the transmission buffers TB1 to TB of its own shared memory M1 to M4.
This can be realized by providing each of the receiving buffers RB1 to RB4 with a receiving area for each partner station, and by limiting the portion of the received data to be taken into each shared memory M1 to M4 to the portion addressed to the own station.

For example, when sending data from station A to station B, select "To B" in the transmission buffer TB1 of station A.
The data of the part indicated by is sent to the reception buffer of station B.
This can be realized by the information transmission devices of the A station and the B station working together to copy the information to the part indicated by "From A" in RB2. N-to-N communication is possible by performing the above processing in all combinations of stations.

For transmission, if the total number of stations is N, then 1, 2, 3...
..., N-1, N, 1, 2..., and the transmission data is the entire contents of the transmission buffer. In other words, if the transmission order is determined, for example, the transmission order is A, B, C, D, A, B, C, D... becomes.

The transmission data includes data indicating the station that has transmitted, and by referring to this data, each station knows that it is its turn to transmit.

On the receiving side, of the received data, only the area allocated to the local station is transferred to the receiving buffer.

When the transmitting station completes one round, each station receives data once from all partner stations.

Here, if synchronization data for each partner station is included in the data exchanged with each other, data transmission can be synchronized in the same way as in the one-to-one transmission described above.

〔Effect of the invention〕

As described above, according to the present invention, the information processing device side updates and updates the synchronization data in the transmitted data without synchronizing the data exchange similar to a handshake method between the information processing device and the information transmission device.
By using references, synchronization is achieved on the information device side, which reduces the number of interfaces between the information processing device and the information transmission device, simplifies both mechanisms, and improves efficiency during high loads. Information transmission can be performed.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the system configuration of an information transmission system in an embodiment of the present invention, Fig. 2 is a flowchart of control operations by the program in Fig. 1, and Fig. 3 is a block diagram showing the system configuration of an information transmission method in an embodiment of the present invention. Block diagram showing the configuration of each shared memory in case 4.
The figure is a block diagram showing the system configuration of a conventional information transmission system. A and B are information processing devices, M1 and M2 are shared memories, TB1 and TB2 are transmission buffers, RB1 and RB
2 is a reception buffer, T is an information transmission device, a and b are communication channels, and l is a transmission path. In addition, in the figures, the same reference numerals indicate the same or equivalent parts.

Claims (1)

[Claims] 1. An information processing device is connected to another information processing device through an information transmission device, the information processing device and the information transmission device exchange data via a first shared memory, and the other information processing device In the information transmission method in which data is exchanged between the information processing apparatus and the information transmission apparatus via a second shared memory, when the information processing apparatus transmits the data to the other information processing apparatus, the information processing apparatus When the data is written to the transmission buffer of the first shared memory and the value of the synchronization data of the transmission buffer is set to a predetermined value, the information transmission device transmits the data to the reception buffer of the second shared memory. At the same time, the value of the synchronization data of the reception buffer is set to a predetermined value, and when the other information processing device detects that the synchronization data has reached the predetermined value, the data is read from the reception buffer. After setting the value of the synchronization data in the transmission buffer of the second shared memory to a predetermined value, the information transmission device transmits the data in the transmission buffer of the second shared memory to the reception of the first shared memory. At the same time, the value of the synchronization data of the reception buffer of the first shared memory is set to a predetermined value, and when the information processing device detects that the value of the synchronization data has reached the predetermined value, a series of transmissions is performed. An information transmission method characterized by terminating processing.