JPH07111697B2 - High-speed communication buffering control method - Google Patents

Description

The present invention relates to a high-speed communication buffering control method using a network to which a processing device is connected.

[Prior Art] A computer, an image terminal, connected via a communication path such as a local area network (hereinafter referred to as LAN),
A system for organically connecting processing devices (hereinafter referred to as nodes) such as a word processor, a workstation, a print server, and a disk server is currently being constructed.

In addition, standardization of each layer has advanced in the LAN, and standardization concepts such as file transfer and job transfer operations are also solidifying.

As a general layered model, as shown in Table 1, the International Organization for Standardization (ISO) open system interconnection or open system interconnection (Open Syste
A reference model called "ms Interconnection" (hereinafter referred to as an OSI reference model) is known.

This OSI reference model consists of seven layers from communication line control to business-dependent communication functions, that is, from the upper layer, in order from application (level 7), presentation (level 6), session (level 5), transport ( Level 4), network (level 3), link (level 3), and physical (level 1) protocols are layered.

FIG. 11 shows a communication procedure at a transmitting station and a receiving station according to such a conventional layered protocol.

At the transmitting station, first, the data 31 transmitted by the protocol of the application layer and the presentation layer 31
Is created.

In the following, the session layer protocol 32 adds the session level header 32a to the data 31 to create the session layer data 32, and the transport layer protocol adds the transport level header 33a to the data 32 to create the transport layer The data 33 is created, and the network level header 34a is added to the data 33 by the protocol of the network layer to add the data 3 of the network layer.
4 is created.

Finally, the data link level header 35a is added to the data 34 by the data link layer protocol to create the data link layer data 35. This data 35 is transmitted to the receiving station via the transmission medium by the interface method defined in the physical layer.

At the receiving station, the data link level header 35a added at the transmitting station, the network level header 34a,
The transport level header 33a and the session level header 32a are sequentially removed in each layer, and the data 31 is reproduced in the application layer and the presentation layer.

Each header 35a to 32a is used as control information in the receiving station. The headers 35a to 32a enable connection with many network systems and provide future interoperability.

In this case, each station is configured so that a plurality of modularized protocol software are linked, and data is actually copied between each layer to store and pass it to the next layer.

[Problems to be Solved by the Invention] However, in the above-mentioned conventional method, the same data is copied in each layer, the storage area of the data increases, and the copy time becomes longer than the protocol processing time.
There is a problem that the execution speed decreases. Further, the header increases from the upper layer to the lower layer, and the original data to be transferred is compressed in the packet.

Furthermore, with regard to the protocol, if the standardization proposals of each layer are completely covered, then multiple classes of protocols must all be implemented. In this case, the software for each class is not actually created, and the same software is used for the common part. Therefore, as the number of classes increases, the process flow of class selection frequently appears, causing the overhead. Becomes

Therefore, in the above-mentioned conventional method, the headers 32a to 3 of each layer are
Since 5a is large, there is a problem that the actual transmission efficiency of the amount of data actually sent to the transmission path is reduced, and the number of layers is large, so that the processing time until communication is long.

In view of the above conventional problems, it is an object of the present invention to improve the effective transmission efficiency of the amount of data actually sent to the transmission path, and to reduce the processing time until the communication. It is to provide a buffering control device.

[Means for Solving the Problem] According to the present invention, the object is a high-speed communication buffering control method having a layered network structure, wherein data of each layer is based on a first management table. If the data is managed, and the data has the same processing and the same header, the data of the application is packet-divided, and the pointer of the packet-divided data indicating the same header is rewritten based on the second management table. Is achieved by a high speed communication buffering control method.

[Operation] Manages data of each layer based on the first management table,
If the data has the same processing and the same header, the data of the application is divided into packets, the pointer of the divided data indicated by the same header is rewritten based on the second management table, and the data is transferred. On the other hand, if the data is not the same process and the same header, the protocol process and the header creation process are performed, the header information is entered in the first management table, and the process moves to the next lower layer.

Embodiments Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 schematically shows data transfer between layers by a high-speed communication buffering control method according to an embodiment of the present invention.

In Fig. 1, n is a positive integer and the data of the nth layer
11 has a header 12 of the (n-1) th layer added (n-
1) The data is transferred to the layer 13, and the header 14 of the (n-2) th layer is added to the transferred data (n-2).
Delivered to layer 15.

FIG. 2 (a) shows the physical structure in the high-speed communication buffering control method of FIG. 1, and FIG. 2 (b) is included in the physical structure of FIG. 2 (a). 1 shows a management descriptor table DT as a table means having a layer pointer 16, a data storage pointer 17 for each layer, and an area 18 as an area for storing a data (header) length.

Next, the physical structure firstly comprises the data 19 of the nth layer of length DLn written from the address An. Hereafter, the length DLn written from address An-1 sequentially from the upper layer
-1, (n-1) th layer header 20, and (n-2) th layer header 21 of length DLn-2 written from address An-2
Etc.

Further, the physical structure includes the pointer (DTn + 1) of the next (n + 1) th layer, the data storage pointer An of the nth layer, and the data length DLn of the nth layer, which are written from the address DTn. The next nth layer pointer DTn written from DTn-1, the (n-1) th layer data storage pointer An-1, the data length DLn-1, the address DTn-
The pointer (DTn) of the next (n-1) th layer written from 2
-1, a (n-2) th layer data storage pointer An-2, a data length DLn-2, and the like, and a management descriptor table DT.

FIG. 3 shows the configuration of each layer from the data link layer as the first layer to the application layer as the second layer.

In the data link layer 22 shown in FIG. 3A, a pointer NP (Next Pointer) to the next network layer 23,
Data storage pointer DP of data link layer and data length DL
And are created.

In the network layer 23 shown in FIG. 3B, a pointer NP to the next transport layer 24, a data storage pointer DP of the network layer, and a data length DL are created.

In the transport layer 24 shown in FIG. 3 (c), the pointer NP to the next session layer 25 and the transport layer
24 data storage pointers DP and a data length DL are created.

In the session layer 25 shown in FIG. 3D, the pointer NP to the next presentation layer 26 and the session layer
25 data storage pointers DP and a data length DL are created.

In the presentation layer 26 shown in FIG. 3 (e), a pointer NP to the next application layer 27, a data storage pointer DP of the presentation layer 26, and a data length DL are created.

Finally, in the application layer 27 shown in FIG. 3 (f), the last pointer FFFFH of the data storage area, the data storage pointers DP1 to DP3 of the application layer 27, and the data lengths DL1 to DL3 are provided for each packet. Is created.

FIG. 4 shows pointers in the presentation layer 26 shown in FIG. 3 (e) to the application layer 28 shown in FIG. 3 (f), and the data pointer of each packet is managed by the data pointer management table 29. .

Therefore, according to the present embodiment, an actual packet is created only by rewriting the pointer NP of the upper layer for each layer, and the data and header of each layer show the connection state and the actual state by the management descriptor table. Can be managed individually.

5 (a) and 5 (b) show a process of creating the management descriptor table DT at the time of transmission.

In FIG. 5 (a), first, the head pointer of the data storage area is acquired from the application (step 41),
A data descriptor pointer DP is obtained for each packet to create a management descriptor table DT (step 42), and a data pointer management table is created using this data pointer pointer DP.
Create 29 (step 43).

The processing of each subsequent layer (step 44) differs depending on the protocol, but as shown in FIG. 5 (b), basically the same header, if the same processing, the header is not created, The pointer NP of the next upper layer of the data storage pointer DP is simply entered in the management descriptor table DT to proceed to the next lower layer (steps 441, 442, 445).

When transmitting a large-capacity divided packet, the above processing is performed to divide only the application layer data.

On the other hand, when the same header and the same processing are not performed, the protocol processing and the header creation processing are performed (step
443), fill in the header information in the management descriptor table DT (step 444), and proceed to the next lower layer (step 44).
Five).

Upon completion of the processing of each layer (step 44), returning to FIG. 5 (a) again, preparation of the packet for transmission is completed, and all data is transmitted (steps 45 to 47). Therefore, the processing can be performed faster than the conventional processing for each layer.

FIG. 6 shows a procedure for transmitting the transmission packet to the transmission medium. When a transmission request is generated from the communication LSI or hardware, the area of each layer of the management descriptor table DT is chained to transmit the packet (step 48). ,
The next packet refers only to the data pointer management table 29 (step 51) and only rewrites the pointer NP of the upper layer of the management descriptor table DT (step 5).
2).

7 (a) and 7 (b) show a process of creating the descriptor table at the time of reception.

As shown in FIG. 7 (a), as in the case of transmission, first, the head pointer of the data storage area is acquired from the application (step 61), the data pointer management table 29 is prepared (step 62), and then , When the packet is received (step 63), the process moves to each layer (step 64, steps 641 to 645 in FIG. 7B).

As for the processing of each layer, basically the same header, if it is the same processing, do not dare create the header,
The pointer NP to the next layer of the data storage pointer DP is simply entered in the management descriptor table DT to advance to the next upper layer (steps 641, 642, 645).

If the same header and the same processing are not performed, the protocol processing and the header creation processing are performed (step
643), write the header information in the management descriptor table DT (step 644), and proceed to the next upper layer (step 6).
45).

When the processing of each layer (step 64) is completed, FIG. 7 (a)
Back, the packet is ready to be received and all data can be received (steps 65, 66).

FIG. 8 shows a procedure for receiving a packet from a transmission medium. When a reception interrupt occurs from a communication LSI or hardware, the header part is cut out from the packet data and stored in the header storage area of each layer (step 67). A data descriptor is stored in the data storage area to create a management descriptor table DT (step 68), and is written in the data pointer management table 29 (step 70).

According to the present embodiment, at the time of reception, only the data is already continuously stored in the data storage area, so that the data storage area of the data storage area can be stored only when the processing of each layer progresses and permutation or retransmission occurs. The inside is operated.

Therefore, it is possible to reduce unnecessary headers and processing of each layer for the application, and by extracting only the data included in the packet, it is possible to reduce unnecessary copying of data between layers and perform high-speed communication processing. The intelligent board system in which the network protocol up to the predetermined layer shown in FIG. 9 is processed by another processor and the communication LSI shown in FIG. 10 support up to the data link layer, and the main CPU Processing can also be performed in a non-intelligent board system that processes upper layers.

In the intelligent board system shown in FIG. 9, the intelligent board which includes simple main transfer protocol (SMTP), file transfer protocol (FTP) and telnet (TELNET) as the application layer, presentation layer and session layer of the upper protocol layers. It is divided into parts of the system, and the transport layer, network layer, and data link layer are the Transmission Control Protocol (TCP), Internet Protocol (I
P), User Datagram Protocol (UDP), Address Resolution Protocol (ARP) and Internet Control Message Protocol (ICMP).
In the non-intelligent board system shown in Fig. 10, each layer from the application layer to the network layer in the upper protocol layer is shared by the host support area of the non-intelligent board system, and the data link layer is the non-intelligent board system. of
It is shared by the LAN board.

[Effect of the invention] Data of each layer is managed based on the first management table,
If the data has the same processing and the same header, the data of the application is divided into packets, the pointer of the divided data indicated by the same header is rewritten based on the second management table, and the data is transferred. Since it is not necessary to copy the data of each layer, the processing time up to communication can be shortened, and the data movement processing can be minimized, so that the data storage area can be minimized and actually It is possible to improve the effective transmission efficiency of the amount of data sent to the transmission path.

[Brief description of drawings]

FIG. 1 is an explanatory diagram showing data transfer between layers in a high-speed communication buffering control device according to an embodiment of the present invention, and FIG. 2 (a) is a high-speed communication buffering control device in FIG. An explanatory diagram showing an actual physical structure, FIG. 2 (b) is an explanatory diagram showing a management descriptor table included in the physical structure of FIG. 2 (a), and FIG.
Explanatory diagrams showing descriptor tables of each layer from the data link layer to the application layer, FIG. 4 is an explanatory diagram showing a pointer of the next application layer in the presentation layer, and FIGS. 5 and 6 show the operation at the time of transmission. 7 and 8 are flowcharts showing the operation at the time of reception, and FIGS. 9 and 10 are explanatory diagrams showing hardware to which the high speed communication buffering control device of the present invention is applied, FIG. 11 is an explanatory diagram showing operations of a conventional transmitting station and receiving station. 11 ... nth layer data, 12 ... (n-1) th layer header, 13 ... (n-1) th layer, 14 ... (n-2) th layer header, 15 ... third layer (N-2) layer, 16 ... Next upper layer pointer, 17 ... Data storage pointer, 18 ... Data length storage area, DT ... Management descriptor table, 29 ... Data pointer management table.

Claims (1)

[Claims] 1. A high-speed communication buffering control method having a hierarchical network structure, wherein data of each layer is managed based on a first management table, and the data has the same processing and the same header. If so, the high-speed communication buffering control method is characterized in that the data of the application is packet-divided, the pointer of the packet-divided data indicating the same header is rewritten based on the second management table, and the data is transferred. .