JPH0420153A - Communication control system - Google Patents

Abstract

PURPOSE:To easily expand the system rich in flexibility by providing means giving chain information representing connection of plural split data split by a data split means to each split data and sending the split data sequentially to a data processing node. CONSTITUTION:The system is provided with a reception means 14 receiving a data of a 1st data length sent from a host computer 5, a data split means 12 spitting the data of the 1st data length received by the reception means 4 into plural data in the unit of a 2nd data length corresponding to a data processing node 1 and a means 13 giving chain information representing connection of plural split data split by the data split means 12 to each split data and sending the split data sequentially to the data processing node 1. Thus, it is not required to set a maximum data length identical able to be sent and received between a host computer 5 and the data processing node 1. Thus, the system rich in flexibility is easily implemented.

Description

[Detailed Description of the Invention] [Purpose of the Invention (Industrial Application Field) This invention relates to a communication control method for a cooperator network system, and particularly to a communication control method for controlling communication between a host computer and a data processing node. .

(Prior Art) In general, in a computer network system equipped with lower-level distributed processing processors that do not have an environment that allows direct connection to a host computer, these lower-level distributed processing processors are used for distributed processing of another higher-level model. Connected to a host computer through a processor.

In this case, the lower model distributed processing processor is called a lower node distributed processing processor, and the higher model distributed processing processor is called an upper node distributed processing processor. The upper node distributed processing processor executes protocol conversion between the lower node distributed processing processor and the host computer in order to realize communication therebetween.

FIG. 6 conceptually shows an example of such a computer network system. The lower node distributed processing processors 1, 2.3 are connected to a plurality of data processing terminals 1a to 1n, 2a to 2n, and 3a to 3n, respectively, and are commonly connected to the host computer 5 via the upper node distributed processing processor 4. It is connected.

In the computer network system of FIG. 6, for example, when transmitting data from the lower node distributed processing processor 1 to the host computer 5, the lower protocol corresponding to the lower node distributed processing processor 1 is the protocol conversion of the upper node distributed processing processor 4. The protocol is converted into an upper protocol compatible with the host computer 5 by the function. In addition, when transmitting data from the host computer 5 to the lower node distributed processing processor 1, the upper level protocol corresponding to the host computer 5 is changed to the lower level protocol corresponding to the lower node distributed processing processor l by the protocol conversion function of the upper node distributed processing processor 4. converted into a protocol.

However, in such a computer network system, if the data length of the data transmitted from the host computer 5 to each of the lower node distributed processing processors 1 to 3 is longer than that of each of the lower node distributed processing processors 1 to 3,
If the data length is larger than the maximum data length that can be sent and received, a problem may arise in which normal data transfer cannot be performed. For this reason,
In order to improve the performance of the system, it is necessary not only to replace the host computer 5 with one of a higher-level model, but also to replace the host computer 5 and each lower node distributed processing processor 1 to
In order to equalize the maximum data lengths that can be transmitted and received in nodes 3 and 3, it is necessary to make a large-scale system change such as replacing each of the lower node distributed processing processors 1 to 3 with higher-level models.

Therefore, conventionally, a system must always be constructed taking into consideration the maximum data length that can be transmitted and received between a host computer and a lower node distributed processing processor, making it difficult to expand the system in a flexible manner.

(Problem to be Solved by the Invention) Conventionally, if the maximum data lengths that can be transmitted and received between a host computer and a lower node distributed processing processor are different, normal communication cannot be executed. The drawback is that the system must be constructed taking into account the maximum data length that can be transmitted and received.

This invention was made in view of the above points, and it is possible to perform normal communication even when the maximum data length that can be sent and received between the host computer and the lower node distributed processing processor is different, thereby providing a highly flexible system. The purpose of this invention is to provide a communication control method that can easily realize system expansion.

[Structure of the Invention] (Means and Effects for Solving the Problems) A communication control method according to the present invention controls communication between a host computer and a data processing node, and the first data length transmitted from the host computer a receiving means for receiving data of a first data length, a data dividing means for dividing the data of a first data length received by the receiving means into a plurality of data in units of a second data length corresponding to the data processing node, and this data dividing means. - Adding chain information indicating the connection of a plurality of divided data divided by - to each of the divided data, and sequentially transmitting the divided data to the data processing node.

In this communication control method, data of a first data length sent from a host computer to a data processing node is divided into a plurality of pieces of data in units of a second data length corresponding to the data processing node, and each divided data tends to chain information is added to indicate the connection of divided data. Therefore, even if the data length sent from the host computer to the data processing node is larger than the maximum data length that that data processing node can send and receive, the data processing node can normally receive data from the host computer. . Therefore, it is no longer necessary to set the maximum data length that can be transmitted and received between the host computer and the data processing node to be the same, and it becomes possible to easily realize flexible system expansion.

(Example) Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1 shows the configuration of a computer system for realizing a communication control method according to an embodiment of the present invention. Similar to the conventional system shown in FIG. 6, this computer system has a configuration in which protocol conversion between the lower node distributed processing processors 1 to 3 and the host computer 5 is executed by the upper node distributed processing processor to control communication between them. However, only the configuration corresponding to one lower node distributed processing processor l is representatively shown here.

The upper node distributed processing processor 10 in FIG. 1 operates under the same protocol as the host computer 5, and performs protocol conversion between the lower node distributed processing processor I and the host computer 5, and also data length conversion between them. Execute.

The protocol conversion unit of this upper node distributed processing processor 10 includes a first line control table 11, a second line control table 12, a chain data transmission buffer 13, a reception buffer 14, A communication interface 15 is provided. Also, the first
Reference numerals 12' and 14' in the figure are a line control table and a receiving buffer provided for realizing a plurality of logical channels, and these line control table 12' and receiving buffer 14' are the line control table 12 and receiving buffer 14', respectively. It has the same configuration as .

The first line control table 11 is logically connected to the lower node distributed processing processor 1, and receives and passes information between the lower node distributed processing processor 1 and the upper node distributed processing processor 10. This first line control table 1
1 is composed of an external interface table lla and an internal interface table llb. In the external interface table lla,
Information indicating the node name of the lower node distributed processing processor 1 and the logical channel name to be connected to the lower node distributed processing processor 1 is registered, and based on this information, the line control table 11 is logically connected to the line control table 12, for example. be done. Further, various types of information for exchanging information between the line control table 11 and the line control table 12 are registered in the internal interface table llb.

The chain data transmission buffer 13 is for transmitting data to the lower node distributed processing processor l by logical combination with the first line control table 11, and has a maximum data length ( For example, it has the same storage capacity as 256 bytes).

The second line control table 12 includes the host computer 5
Information is received and passed between the host computer 5 and the upper node distributed processing processor 10 via the communication interface 5. This second line control table 12 is an external interface table 12.
a and an internal interface table 12b.

Registered in the external interface table 12a are the logical channel number names assigned to the line control table 12 and information indicating the node names to be connected to the host computer 5. Based on these information, the line control table 11 is It is logically coupled to the line control table 12.

The internal interface table 12b also contains various information for exchanging information between the line control table 12 and the line control table 11, as well as segmentation information for dividing data sent from the host computer 5 into multiple pieces. A pointer (S P) 121 and a protocol conversion execution unit (PC) 122 that adds control information to each divided data are provided.

The reception buffer 14 is for receiving data transmitted from the host computer 5 through logical connection with the second line control table 12, and has a maximum data length that can be transmitted and received by the host computer 5 (for example, 4 Mbytes). ) has the same storage capacity.

Next, the principle of the data length conversion operation executed by the segmenting pointer 121 of the line control table 12 and the protocol conversion execution unit 122 will be explained with reference to FIGS. 2 to 4.

FIG. 2(A) shows a data format when the maximum data length that can be transmitted and received by the lower node distributed processing processor l is 256 bytes. As shown in the figure, the first 9 bytes of the 256 bytes are a header section consisting of control information for transmission or reception, and the remaining 247 bytes are a data section consisting of actual transmission or reception information.

FIG. 2(B) shows a data format when the maximum data length that can be transmitted and received by the host computer 5 is 4 Mbytes. As shown in the figure, the first 9 bytes of the 4 Mbytes are a header section consisting of control information for transmission or reception, and the remaining 4083 bytes are a data section consisting of actual transmission or reception information.

The segmenting pointer (SP) 121 is shown in FIG.
4M bytes of data transmitted from the host computer 5 as shown in B) is divided as shown in FIG. That is, 4 M transmitted from the host computer 5
Of the byte data, a 9-byte header section is stored in the protocol conversion execution unit 122, and the remaining 4083-byte data section is stored in the reception buffer 14. and,
The 4083-byte data portion is divided by the segmenting pointer 121 into a plurality of divided data D1 to Dn each consisting of 247 bytes.

These 247 bytes correspond to the size of the data portion of the data handled by the lower node distributed processing processor 1 shown in FIG. 2(A), which is 247 bytes.

The divided data D1 to Dn are sequentially stored in the chain data transmission buffer 13. In this case, the divided data Di-D
The protocol conversion execution unit 122 adds 9 bytes of control information to each n as a header. This control information includes information indicating the connection between the divided data D1 to Dn.

For example, as shown in FIG. 4(A), part of the 9-byte control information added to the divided data D1 includes chain information (1,0 ) is set.

In addition, as shown in FIG. 4(B), part of the 9-byte control information given to each of the intermediate divided data D2 to Dn-1 between the divided data DI and the divided data Dn includes the divided data Chain information (0, 0> is set to indicate that D2 to D n-1 are intermediate data.Furthermore, part of the 9-byte control information given to the divided data Dn includes the information shown in FIG. As shown in (C), chain information (0, 1) indicates that the divided data Dn is the last data.
is set.

Next, the operation of the protocol conversion process executed by the upper node distributed processing processor 1o shown in FIG. 1 will be described with reference to the flowchart in FIG.

The upper node distributed processing processor 10 first checks the data length of the data received from the host computer 5, and determines whether the data length is larger than the maximum data length of 256 bytes that can be transmitted and received by the lower node distributed processing processor. (Step AI). If the received data is 256 bytes or less, the received data stored in the receive buffer 14 is sent as is to the lower node distributed processing processor 1 via the chain data send buffer 13 (step A2), and then a receive request is sent to the host computer 5. (Step A3).

On the other hand, the data received from the host computer 5 is 256
If it is larger than the byte, 9 bytes of control information (header part) is extracted from the received data and stored in the protocol conversion execution unit 122 of the line control table 12 (step A4).

Next, the received data stored in the receive buffer 14 is
The first 247 by segmenting pointer 121
The byte is divided as the first divided data Di, and the divided data DI is stored in the chain data transmission buffer 13 together with 9 bytes of control information including chain information (1, 0), and then sent to the lower node distributed processing processor 1. It is transmitted (step A5). The value of the segmenting pointer 121 is then moved to the last position of the transmitted data.

Thereafter, the upper node distributed processing processor 10 checks the data length after the value specified by the segmenting pointer 121, that is, the remaining data length in the reception buffer I4, and the remaining data length is the one that can be transmitted and received by the lower node distributed processing processor. It is determined whether the maximum data length is greater than 256 bytes (step A6). Remaining data length is 25
If it is larger than 6 bytes, the segmenting pointer 121 divides the remaining data in the receive buffer 14 again as 247-byte intermediate divided data D2, and the divided data D2 is 9 bytes including chain information (0,0). control information is stored in the chain data transmission buffer 13, and transmitted to the lower node distributed processing processor 1 (step A7). The value of the segmenting pointer 121 is then further moved to the last position of the transmitted data.

The processes of steps 80 to A7 are repeatedly executed until the data length after the value specified by the segmenting pointer 121, that is, the remaining data length in the reception buffer 14 becomes 256 bytes or less.

When the remaining data length of the receive buffer 14 becomes 256 bytes or less, the remaining data after the value specified by the segmenting pointer 121 is taken out from the receive buffer 14 as the last divided data Dn, and the last divided data Dn is stored in the chain data transmission buffer 13 together with 9 bytes of control information including chain information (0, 1), and is transmitted to the lower node distributed processing processor 1 (step A8).

As explained above, in this embodiment, the maximum 4M bytes of data sent from the host computer 5 to the lower node distributed processing processor 1 is divided into multiple pieces of data in data length units corresponding to the lower node distributed processing processor 1. The data is divided, and each divided data is given chain information to indicate the connection of the divided data. Therefore, no matter how large the data length transmitted from the host computer 5 is, the lower node distributed processing processor 1 can normally receive the data from the host computer 5.

[Effects of the Invention] As described above, according to the present invention, there is no need to set the same maximum data length that can be transmitted and received between the host computer and the data processing node, and flexible system expansion can be easily achieved. becomes possible.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a system configuration for realizing a communication control method according to an embodiment of the present invention, and FIG. 2 shows a host computer and a data processing node provided in the system shown in FIG. Figure 3 is a diagram for explaining the data format of the corresponding transmitted and received data, Figure 3 is a diagram for explaining the data division operation performed by the system shown in Figure 1, and Figure 4 is a diagram for explaining the data division operation performed by the system shown in Figure 1. Figure 5 is a flowchart explaining the protocol conversion operation executed in the system shown in Figure 1. Figure 6 is the system configuration of the system shown in Figure 6. FIG. DESCRIPTION OF SYMBOLS 1... Lower node distributed processing processor, 5... Host computer, 10... Upper node distributed processing processor, 11.12... Line control table, 13...
Chain data transmission buffer, 14... Reception buffer, 121... Segmenting pointer, 122...
・Protocol conversion execution unit.

Claims (1)

[Scope of Claim] A communication control method for controlling communication between a host computer and a data processing node, comprising: receiving means for receiving data of a first data length transmitted from the computer; and a first data length received by the receiving means. data length data to the second node corresponding to the data processing node.
A data dividing means that divides the data into a plurality of pieces of data in units of data length; and chain information indicating the connection of the plurality of pieces of divided data divided by the data dividing means to each piece of divided data; 1. A communication control method, comprising: means for sequentially transmitting data.