JPS61264945A - Protocol parallel processing system of communication processor - Google Patents

Abstract

PURPOSE:To decrease the processing time per data unit by giving directly data link protocol control information, other N protocol control information and remaining information to a corresponding processing program respectively. CONSTITUTION:A distribution processing execution program 12 in a data link layer processing processor 6 receiving a data link data unit 15 from a physical layer processing processor 5 disconnects a data link header 151, a trailer 152, a network header 153, transport headers 154a, 154b and gives directly them to a corresponding processing processor respectively. Further, the remaining part, that is, a section data unit 18 is given directly to a session layer processing processor 9. Thus, the protocol processing of each layer is started without awaiting the end of the processing of the lower layer and the total processing time per data link data unit is decreased.

Description

[Detailed Description of the Invention] [Field of Industrial Application] This invention relates to a protocol for a communication processing device that is composed of a plurality of microprocessors interconnected by a system bus and executes a communication protocol consisting of a plurality of layers. It is related to parallel processing methods.

[Conventional technology]

FIG. 1 is a block diagram illustrating one embodiment of the present invention, but FIG. The figure will be explained as a configuration of a conventional device.

In the figure, l is a computer, 2 is a data communication network, 3 is a communication processing device that is interposed between the computer l and the data communication network 20 and processes the communication protocol, and 4 to 1O are the components of the communication processing device 3. 4 is a system bus, 5 is a physical layer processor, 6 is a data link layer processor, 7 is a network layer processor, 8 is a transport layer processor, 9 is a session layer processor, and 10 is a computer interface processor. It is.

At times, the physical layer processing processor 5 sends and receives data to and from the data communication network 2, and the computer interface processor 1
0 is for transmitting and receiving data with the computer.

FIG. 6 is a diagram showing the flow of conventional processing for received data input from the data communication network side within the communication processing device having the configuration shown in FIG. 1. Be busy, 5.
10 is the same processor as in FIG. 1, 15 is a data link data unit which is N data units of the data link layer, and 151 and 152 are data link headers and data which are protocol control information of the data link layer, respectively. Link trailer, 153, network header, which is network layer N protocol control information, 154m
, 154b is a transport header that is N protocol control information of the transport layer, 155 is a session header that is N protocol control information of a session layer, 156
is session-user data, which is data communicated between programs above the session layer, and 16.17.18 is network data unit, which is N data units of the network layer, transport layer, and session layer, respectively. Data unit, session data unit.

Next, the operation will be explained. First, data transferred via the data communication network 2 is received by the physical layer processor 5 and passed to the data link layer processor 6. However, the transfer of data within the communication processing device 3 does not necessarily involve physical movement of the data;
For example, processing such as passing the start address of the data storage area may be used.

The data link layer processing processor 6 that has received the data link data unit 15 processes it according to the N protocol control information 151, 152 of the data link layer, and if there are remaining parts 153 to 156, the network e data unit 1
6 to the network layer processing processor 7. After receiving the network data unit 16, the network layer processing processor 7 performs processing according to the N protocol control information 153 of the network layer, as in the case of the data link layer described above, and then processes the remaining portions 154a to 156. If so, it is passed to the transport layer processor 8 as a transport data unit 17.

After that, the transport layer and the session layer are processed in the same way, and finally, the session user data 156
If any remains, it is transferred to the computer 1 through the computer interface processor 10.

[Problem that the invention seeks to solve]

Processing of received data in a conventional communication processing device is carried out according to the above-mentioned procedure. Therefore, regarding the reception processing of one data link data unit, the upper layer processing processor finishes the processing of the lower layer. There was a problem that the total processing time per data link data unit was large. This invention was made to solve the above problems, and the processing of the upper layer It is an object of the present invention to obtain a protocol parallel processing method for a communication processing device that shortens the waiting time of a processor and shortens the total processing time per data link data unit.

[Means for solving problems]

In the protocol parallel processing method of the communication processing device according to the present invention, in addition to the means in which the data link layer processing processor executes the normal data transmission control procedure, the data link layer processor analyzes the received data link data unit and The N protocol control information up to the top layer to be distributed is separated, and the data link protocol control information is sent to the program that executes the data transmission control procedure, and the remaining N protocol control information is sent to the processing program of the corresponding layer. The parts are directly handed over to the processing programs in the layer one level higher than the top layer to be distributed.

[For production]

The protocol parallel processing method of this invention makes it possible to start processing N protocol control information from the data link layer to the layer one layer above the topmost layer subject to distribution processing without waiting for the completion of processing in the lower layers. , the upper layer latency is reduced, and the total processing time per data link data unit is reduced.

〔Example〕

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

FIG. 1 is a block diagram showing one embodiment of the present invention,
The structure and the method of sequentially processing from the lower layer are as described above. Further, FIG. 2 is a block diagram showing the internal configuration of the data link layer processing processor 6. As shown in FIG. In the figure, 11 is a data transmission control procedure execution program, and 12 is a data transmission control procedure execution program that analyzes the received data link e data unit, separates N protocol control information from the data link layer to the highest layer to be distributed, and separates the N protocol control information including the remaining parts. A distribution processing execution program that executes distribution processing directly handed over to a corresponding processing processor or program, 1
3 is whether or not the process is in the process of assembling N data units by combining data that has not been completed as N data units with the related incomplete N data units that have already been received or the related incomplete N data units that will be received from now on. An assembly flag table 14 is a work area for holding received data during distribution processing and the above-mentioned N incomplete data units.

FIG. 3 shows the flow of processing for data received from the data communication network side when the protocol parallel processing method according to the present invention is used inside the communication processing device configured as shown in FIGS. 1 and 2. It is something that Here, the transport layer is the highest layer to be distributed, and protocol processing from the data link layer to the session layer is carried out in parallel.

All numbers in the figure are the same as those in FIG.

Next, the operation will be explained. A distribution processing execution program 12 in the data link layer processing processor 6 that has received the data link data unit 15 from the physical waste processing processor 5.
separates the data link header 151, trailer 152, network header 153, and transport headers 154a+, 154b according to the algorithm described below, and directly delivers them to the corresponding processors.

Further, the remaining portion, that is, the session data unit 18, is also directly delivered to the session layer processor 9. The data transfer from the session layer processor 9 to the computer interface processor 10 is the same as in the conventional system.

4 and 5 are flowcharts showing the execution algorithm of the distribution process. 401-417,5 in the figure
01 to 517 are each processing step. Here, HD L is used as a data transmission control procedure in the data link layer.
C (High-1 level DataLink Co
X.25 packet level protocol is adopted as the network layer protocol. In the figure, NRF, TR, and F are assembly flags for the network layer and transport layer, respectively, and are stored in the assembly flag table 13. When the value of the assembly flag is 1, it means that N incomplete data units are held, and when the value of the assembly flag is 00, it means that no data units are held. Also, in the figure, NL
, TL, and 8L refer to a network layer processor 7, a transport processor 8, and a session layer processor 9, respectively.

First, in step 401 when starting up the system, N
OK initialize RF and TRF. Next step 402
The following is the HD corresponding to data link data unit 15.
Regarding the processing for LC frames, the process returns to step 402 every time the processing for one HDLC frame is completed.

Step 403 involves cutting out the data link header 151 and trailer 152 and passing them to the data transmission control procedure execution program 11. As a result of step 403, if there is a remaining portion, it is a packet that is a network data unit, so the process proceeds to step 405.

Steps 405 to 417 are processes related to this packet, and in any one of steps 407, 409, and 410, a packet header corresponding to the network header 153 is passed to the network layer processing processor 7. In addition, in the case of a D T (Data) packet responsible for normal data transfer, processing steps 413 to 413 regarding assembling transport data units are performed according to the information in the packet header (M bit in the figure) and the value of NRF. 417 is performed.

As a result of steps 405 to 417, the processing from step 501 onward is performed on data that is completed as a transport data unit. Steps 501 to 516 are processes related to transport data units, and in any of steps 504, 505, 508, and 509, transport headers 154m and 154b are passed to the transport layer processing processor 8. The processing algorithm is almost the same as the processing steps 405 to 417 for packets, except that a plurality of transport headers 154
Step 50 so that m, 154b can be separated one by one.
The parts 1 to 504 form a loop.

Finally, in step 517, the completed session data unit is passed as is to the session layer processor.

〔Effect of the invention〕

As described above, according to the present invention, the data link layer processing processor receives N protocol control information from the data link layer to the highest layer to be distributed, which is included in the received data link data unit, and N protocol control information to the highest layer to be distributed. Since the remaining portion, which is N data units of the layer one layer above the upper layer, can be directly delivered to the processor or program of the corresponding layer, the protocols of each of the above layers can be transferred directly to the processor or program of the corresponding layer. This has the effect of reducing the total processing time per data link data unit.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram showing the internal configuration of the data link layer processor integrated in FIG. 1, and FIG.
4 and 5 are flowcharts showing the execution algorithm of distribution processing, and FIG. 6 is an internalization of the communication processing device with the configuration shown in FIG. 1. FIG. 2 is an explanatory diagram of conventional transmission of received data units. In the figure, 1 is a computer, 2 is a data communication network, 3 is a communication processing device, 4 is a system bus, 5 is a physical waste processing processor, 6 is a data link layer processing processor, 7 is a network layer processing processor, and 8 is a transport layer 10 is a computer interface processor; 11 is a data transmission control procedure execution program; 12 is a distribution processing execution program; 13 is an assembly flag table; 14 is a work area; 15 is data link data. Unit, 151 is data link header, 1
52 is a data link trailer, 153 is a network header, 154 is a frame, 154b is a transbot header, 155 is a session or header, 156 is session user data, 16 is a network data unit, 17 is a transbot data unit , 18 is a session data unit. In the drawings, the same reference numerals indicate the same or corresponding parts.

Claims (1)

[Claims] In a protocol parallel processing method for a communication processing device in which a plurality of microprocessors are interconnected via a system bus, processing of a multi-layered communication protocol is executed by the microprocessor, and the lowest layer of the communication protocol is As a layer, the Nth layer from this first layer is the (N)th layer.
layer and transfer (N) protocol control information within this (N) layer to execute the protocol of this (N) layer, and this (N) protocol control information corresponds to (N+1) data units. (N) The above (N
) layer is the (N) data unit, the data transmission control procedure in the communication protocol is executed in the data link layer, and this data link layer is (N) data unit.
When layered, (N) data units and (N) protocol control information are data link units, data links, and protocol control information, and a specific layer above the data link layer is the highest layer to be distributed, (N) protocol control information of the data link layer or higher is sequentially added to the data passed from the upper layer of the communication protocol to form a data link data unit, and the protocol processing processor of the data link layer normally In addition to the execution program that executes the data transmission control procedure, in addition to the execution program that executes the data transmission control procedure, during the reception process of the data link data unit, the completed (N) data for each layer from the data link layer to the top layer to be distributed. If the above (N) protocol control information is separated from the unit one by one and the above (N) user data remains, this (N) user data is separated from the above (N)
+1) Determine whether the data unit is complete or not, and if the result of the determination is (N+1) that the data unit is not complete, the related uncompleted (N+1) that has already been received.
) data units and related unfinished data units to be received (N
+1) By assembling the (N+1) data units by combining them with the data units, the received data link
The (N) protocol control information included in the data unit from the data link layer to the topmost layer subject to distribution processing is separated, and the data link protocol control information is added to the program that executes the data transmission control procedure. (N) The communication processing device is characterized in that the protocol control information (N) is directly delivered to the processing processors of the corresponding layers, and the remaining part is delivered directly to the processing processors of the layer one layer higher than the topmost layer to be distributed. protocol parallel processing method.