JPH01194547A - Multiplex communication control device - Google Patents

Abstract

PURPOSE:To execute communication with the maximum number of devices even on a compact computer by executing a simultaneous processing for plural messages with the same device to be generated by an asynchronous processing and causing a message buffer corresponding to the respective devices, to be minimum. CONSTITUTION:A transmitting and receiving buffer 24 is shared by a message index 21 and a packet index 22 and transmission and reception with the respective devices are controlled and executed for the unit of the packets. Accordingly, the transmission and reception can be executed without preparing the transmitting and receiving buffer 24 in each device and each message. Since the plural messages can be stored in each device, asynchronous communication, which does not depend on the processing ability of the subject deice, can be realized. Thus, even in the compact computer, the communication can be executed with the maximum number of the devices.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to communication control with a plurality of devices using a LAN, and in particular, multiplex communication control suitable for a communication control device of a central processing unit that communicates with these devices. Regarding.

[Conventional technology]

In conventional devices, as described in Japanese Patent Application No. 60-271538, a buffer for one message is provided for each device connected to a LAN. This message buffer was managed as a packet for sending and receiving one message on the LAN, and provided equal communication services to each device connected to the LAN.

[Problem to be solved by the invention]

The above-mentioned conventional technology does not take into account simultaneous requests for multiple messages to the same device for messages corresponding to each device, and cannot make simultaneous requests to the same device from multiple business processes. Furthermore, in asynchronous simultaneous processing, the processing capability depends on the communication control capability of the device connected to the LAN. In addition, since this message buffer is provided for each device, no consideration is given to the increase in the number of devices, and as the number of buffers increases with the increase in the number of devices, it is In support, problems such as insufficient storage capacity arise.

The present invention is based on a patent that is characterized by simultaneous processing of multiple messages with the same device that occurs through asynchronous processing, and also minimizes the message buffer for each device so that even a small computer can communicate with the maximum number of devices. It's about doing.

[Means to solve the problem]

The above purpose is to standardize the sending and receiving buffer for each device that the multiplex communication control device communicates with, to divide this sending and receiving buffer into packet units, which are sending and receiving units, and to have an empty buffer management flag for each packet, and to send and receive messages. One or more constituent packets are stored and managed using an empty buffer management flag. In addition, we provide a packet index that stores and manages this storage location on a message-by-message basis, and a device-compatible message index that can store and manage multiple messages for each communication target device. This is achieved by accepting transmission and reception within the range of the number of buffers.

[Operation] The multiplex communication control device performs transmission and reception by sharing a transmission and reception buffer with each device using a message index and a packet index, and controlling transmission and reception with each device in units of packets. As a result, a sending/receiving buffer can be sent and received without having to prepare a sending/receiving buffer for each message for each device, so even a computer with a small storage capacity can communicate with a maximum number of devices. or,
Since multiple messages can be stored for each device, asynchronous communication that does not depend on the processing capacity of the target device can be realized, and the computer processing capacity of the device installed can be utilized to the maximum.

〔Example〕

An embodiment of the present invention will be described below with reference to FIGS. 1 to 5. FIG. 2 shows the configuration of this device. 1
indicates a multiplex communication control device, 2 indicates a logical device, and 5 indicates a logical device.
Corresponding to the physical device 2' connected to the network area of , the business process 6 becomes the target device for message communication. Reference numeral 7 is for making a recovery request after the occurrence of abnormality 1, and is for making a recovery request for each logical device or physical line, and the recovery process is performed by the communication control process 3. 8 processes the abnormality stored in 3 on the user side, and when an abnormality occurs, the abnormality information is acquired through the interface processing of 4 upon startup from 3. Here, 2′
One or more packets constituting the message sent from is received by 3, stored in 2 corresponding to 2', and received by newly determined 6 by receiving the final packet constituting the message. to start. 6, which has been activated for reception, can perform reception processing from 2' by taking in the message via the earnest logic devices 2 to 4. Further, when requesting transmission from the business process 6 to 2', the message to be sent and the logical device number are requested from 3 via 4.

In step 4, the destination node port is converted from the logical device number via a predetermined conversion table, and the message is divided into packets based on the message length, edited into transmittable packets, and stored in step 2.

In step 3, the bucket 1- of the message to be transmitted is sent to 5 according to the transmission type when the order of transmission requests to the transmitter comes. In response to a normal response from 5, the next logical device's transmission message is searched in order, and a transmittable packet is extracted and sent to 5. In this way, by sequentially sending the transmittable packets to 5 for each logical device, the intended message is transmitted. Here, 2 has conventionally been realized on a general-purpose computer with a large memory capacity, having a buffer for one message for each logical device.

This is shown in Figure 1 because of the need to perform asynchronous communication on the control computer, which is relatively limited in memory capacity, relatively unrestricted in the number of logical devices, and not dependent on the processing capacity of 2'. We have invented and realized a buffer management method shown below. FIG. 1 shows a buffer management method corresponding to the logical device 2 in FIG. Reference numeral 21 indicates a message index for each logical device, and 22 indicates a packet index for storing a storage location of a packet constituting a message for each logical device. Reference numeral 24 is an area for storing a packet, which is a unit for transmitting and receiving messages via 5 between 3 and 2' in FIG. 2, and is a buffer shared by each logical device. Further, 23 is a flag table that stores the free status of 24. 23 is set when packet 1- is stored in 24. Incidentally, the reset of 23 is collectively set in accordance with the release of the messages of 22 and 21 when all the packets constituting the message are taken out. This is to enable recovery in units of messages when a failure occurs. 31 is.

It stores the current sending completed logical device number for controlling the sending order when sending packets from 3 to 5. Here, FIG. 3 shows an example in which a plurality of packets constituting a message are received by 3 from each device 2' via 5, and the operations of 3 and 2 will be described. In this example, each device 2
In order to identify PL-1, 2, 3, PJ 1, 2 . Let it be Ph-1, 2.

The timing at which these packets are taken in by 3 via 5 is defined as P+1. PJ 1. Let Ph −1°P+−2, . . . Also, at this time, it is assumed that no message exists in 2. Here, 3 is 5
When a packet is received by an interrupt from , and it is successfully taken in, a logical device number is determined from the source address in the packet using a predefined conversion table. If it cannot be determined, the abnormality is reported to 8 as an abnormal packet reception, and also to the device corresponding to the source address.

Also, the abnormal reception buckets 1~ are read out. On the other hand, for normally received packets, the message index corresponding to the logical device number is determined from the empty state of 21, and
1 is stored, 22 is reserved, and the message number of the packet is determined. This message number is determined by the sequence number and identification code that indicate the packet position within the message in a part called the packet ladder located at the front of the packet, allowing the destination device to determine whether it is the first, middle, or last packet. As a result, it is determined only when the first packet is received. Also, at this time, it memorizes that the message is being received and makes it possible to receive the subsequent message packet. Furthermore, in step 3, the storage area for storing the packet is stored in the free area 24, which is a shared storage area with each logical device, while checking the free space management status of 23, and this storage location is stored in 22. do. 22 further stores the number of these packets. In this way, the packets constituting the message from each device are received and stored one after another in 24, and from 3 to the predefined 6 by receiving the final packet constituting the message of the logical device in question. A reception activation request is issued. In this way, the received packets are sequentially stored via 23 in 24 while being managed for free space.
1.22 manages the order in which messages are configured, so that the message storage area that requires the largest storage area is prepared for the maximum number of buckets (i) that can be generated simultaneously from multiple devices. This eliminates the need to prepare the maximum number of devices for the maximum number of messages. In addition, in 3, reception is possible as long as 22 and 24 are free, and if these are full, the device will report that they are full to the device corresponding to the source address, and the device will check the availability. Assuming that the protocol allows retransmission until the end of the message, this method will work as long as there is a storage area for at least the maximum number of packets constituting one message. On the other hand, packet 1-, which is sent by 3 to 2' via 5, is taken out in a manner opposite to that at the time of reception described above. Using the example of FIG. 3, P+-1, PJ-1, Pk 1. Pt
2. ... are taken out from 24 and sent to 25 in this order.

At this time, the message number of 21 and 22 and the corresponding packet management information are held together with 23 and 24 until the final packet of the message is sent normally, and when it is confirmed by 3 that the message has been sent normally. , 3. This is to enable recovery and retransmission of messages in units of messages when an abnormality occurs. In addition, this method of sending data to 5 is particularly effective when the interface between 5 and 2' is low-speed, and when 5 has a storage area in packet units, the communication service to each 2' is can be provided effectively. Next, Figure 4 shows 21.22
This is a description of the process in step 4 for the device 6 to receive a message composed of 24 buckets 1 to 24 managed by the group. Also, Figure 5 shows the message from 6 at 21
．． It is stored as 24 packet groups managed by 22, and 3
This describes the processing in step 4 for sending to each 2'. In FIG. 4, in response to a request from 6, a received message from 2' corresponding to the logical device is retrieved, and this process is positioned at 4. In this process, 2
According to the message order stored in 21, the packets stored in 4 are extracted from 24 into packets constituting a message by 22, message information in the packet is extracted and edited sequentially against the message, and linkage is performed in 6. , managed the message, 21.2
2, 23, and 24 are each set to be vacant. On the other hand, the fifth
In the figure, in response to a message transmission request from 6, the state of the sending optical logic device is checked, and the message from 6 is accepted when the transmission reception is enabled. The accepted message is divided into packets, which are transmission units, and packet header information for transmission is added to the packets.
Each piece of information is stored in , and a transmission request is made to 3. In this way, messages can be exchanged between 6 and 2' using 1.

〔Effect of the invention〕

According to the present invention, the packet storage area corresponding to each logical device can be minimized in the entire multiplex communication control device, resulting in the following effects.

(1) The utilization efficiency of the packet storage area within the multiplex communication control device can be improved.

(2) According to (1) above, it is only necessary to have a maximum packet storage area that constitutes at least one message, so that economical efficiency can be improved by downsizing the device.

[Brief explanation of the drawing]

FIG. 1 is a relationship diagram of information management in a message area for a logical device in a multiplex communication control device according to an embodiment of the present invention, FIG. 2 is a schematic diagram of the overall configuration including FIG. 1, and FIG. 3 is a packet reception timing chart. Figure 1 is a diagram showing a specific usage method, Figure 4 is a message reception processing flowchart in Figure 1, and Figure 5 is a diagram showing how to use Figure 1.
It is a message transmission processing flowchart in the figure. DESCRIPTION OF SYMBOLS 1... Multiplex communication control device, 2... Logical device message area, 3... Communication control processing, 4... Interface processing group, 5... Network area, 21... Message index, 22. ...Packet index, 23...Buffer free management flag, 24...Transmission/reception buffer.乎 2 口・l 4th Ward ~ 3 Prisoners

Claims (1)

[Claims] 1. A central processing unit, one or more devices that communicate with the central processing unit, a LAN that connects these devices, and a communication control device within the central processing unit that communicates with these devices and one or more business processes within the central processing unit. In a communication control device consisting of 1
1. A multiplex communication control device comprising: a message virtual buffer corresponding to one or more devices that communicate with one or more business processes; and a communication control device that multiplexes the virtual buffers.