JPS631632B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION TECHNICAL FIELD The present invention relates to an inter-processor connection system in a data processing system.

Prior Art Conventionally, in order to transfer large amounts of data between multiple processors, the processors are connected using high-speed adapters. A drawback of high-speed adapters is that they must be developed for each different type of processor to be connected.

OBJECTS OF THE INVENTION It is an object of the present invention to provide an inter-processor connection system that overcomes the above-mentioned drawbacks.

In an information processing system configured such that a management processor and a processing processor are connected to a data storage device, and the management processor and the processing processor are connected via a connection device, the management processor converts the first data to be processed into the first processing target data. Pointer information containing processed data storage area information indicating an area for storing the first processed data and second processed data to be processed by the processing processor is written to the data storage device. writing, processing request information including storage area information of the pointer information on the data storage device is transmitted to the processing processor via the connection device, and the processing processor writes the pointer information based on the processing request information. is read from the data storage device, the first data to be processed is read from the data storage device based on the data storage area information in the pointer information, and the first data to be processed and the data to be processed are It is characterized in that the second data to be processed in the pointer information is processed by the processing processor.

Embodiments of the Invention Next, the present invention will be described in detail with reference to the drawings. Referring to FIG. 1, one embodiment of the present invention includes a magnetic disk device 4, a management processor 1 and a processing processor 2 connected to the magnetic disk device 4, and a processing unit that is controlled by a communication line protocol and processes with the management processor 1. It consists of an in-house connection device 3 that connects to a processor 2.

Next, the operation of the present invention will be explained in detail with reference to the drawings.

The management processor 1 shown in FIG. 1 stores one or more first data files in the format shown in FIG. 4 in the magnetic disk device 4 shown in FIG. 1 via the bus cable 41. Here, "c ₁₁ t ₁₁ " in FIG. 4 represents the storage start cylinder number c ₁₁ and track number t ₁₁ on the magnetic disk drive 4 in FIG. 1, and "c ₁₂ t ₁₂ " represents the storage end cylinder number c ₁₂ . The track number t ₁₂ is shown, and "r' ₁₁ , r' ₁₂ . . . r' _1j " means the record number.

A program (not shown) on the management processor 1 of FIG. 1 writes pointer information in the format shown in FIG. 3 to the magnetic disk device 4 via the bus cable 41. Here, "CT" in FIG. 3 is the storage start cylinder number C on the magnetic disk device 4 in FIG.
and track number T, "C'T'" means storage end cylinder number C' and track number T', and "r ₁ ,"
r ₂ ..., r _o ” means a record number.

Next, the program on the management processor 1 shown in FIG. 1 sends processing request information in the format shown in FIG. 2 to the processing processor 2 via the in-house connection device 3 shown in FIG. The program (not shown) on the processing processor 2 of FIG. 1 that receives the processing request information shown in FIG. 2 is a code in which the first 4 bytes I of the processing request information shown in FIG. It recognizes that the following 8 bytes are the start cylinder number C and track number T of the processed data storage area information shown in FIG. Next, the program on the processing processor 2 shown in FIG _{. 1} reads the record r ₁ shown in FIG. recognize,
Cylinder number c ₁₁ of the magnetic disk device 4 in Fig. 1
Then, from track number _t11 , record _r'11 shown in FIG. 4 is read. The program on processing processor 2 uses the processing parameters in record r ₁ in Figure 3.
Based on p ₁ , the second processed data d ₁ and the first processed data d' ₁₁ in the record r' ₁₁ shown in FIG. 4 are processed, and the processing results are sent to the processing processor 2 shown in FIG. save. Next, the program on the processing processor 2 reads record r' ₁₂ shown in FIG. 4 from the magnetic disk device 4 and performs the same processing as record r' ₁₁ . This process is continued until the cylinder number and track number on the magnetic disk device 4 become the same as the cylinder number c ₁₂ and track number t ₁₂ which are the end address of the processed data storage data of record r ₁ shown in FIG. When the processing for record r ₁ in FIG. 3 is completed, the processing results are the second processed data number m ₁ in FIG. 3 and the first processed data numbers m' ₁₁ , m' ₁₂ , etc. shown in FIG. 4. , m' _1j (not shown) from the program on the processing processor 2 of FIG. 1 to the management processor 1 via the in-house connection device 3.

Next, the program on the processing processor ₂ shown in FIG. 1 reads the record r ₂ shown in FIG. Processing is performed in the same manner, and the processing results are sent to the management processor 1 via the in-house connection device 3 shown in FIG.
Processing and sending of processing results regarding records r ₁ to _{r o} in FIG. 3 are performed by changing the storage cylinder number and track number of the record shown in FIG. 3 to the end cylinder number C' and track number of the processing request information shown in FIG. 2. By continuing until they match T', the processing in the processing processor 2 requested by the management processor in FIG. 1 is completed.

By connecting the management processor 1 and the processing processor 2 using the device cross-call function of the magnetic disk device 4 shown in FIG. 1 and the in-house connection device, and using the pointer information shown in FIG. This has the advantage that it is possible to construct an information processing system that connects processors without developing a new interprocessor adapter.

Effects of the Invention The present invention eliminates the need to develop a new device for inter-processor coupling by performing inter-processor coupling using a data storage device such as a magnetic disk device and a connection device such as an in-house connection device. There is an effect.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the configuration of the information processing system of the present invention, FIG. 2 is a diagram showing the format of processing request information, FIG. 3 is a diagram showing the file format of processed data storage area information, and FIG. 1 is a diagram showing a file format of first processed data; FIG. In the figure, 1... management processor, 2... processing processor, 3... in-house connection device, 4... magnetic disk device, 41, 42... bus cable.

Claims (1)

[Scope of Claims] 1. An information processing system in which a management processor and a processing processor are connected to a data storage device, and the management processor and the processing processor are connected via a connection device, wherein the management processor is connected to a first processor. Write the data to be processed into the data storage device, and provide pointer information including data to be processed storage area information indicating an area for storing the first data to be processed and second data to be processed to be processed by the processing processor. writing in the data storage device and transmitting processing request information including storage area information of the pointer information to the processing processor via the connection device, and the processing processor converts the pointer information to the processing request information based on the processing request information. After reading from the data storage device, the first processed data is read from the data storage device based on the processed data storage area information in the pointer information, and the first processed data and the pointer are An inter-processor connection method characterized in that the second data to be processed in the information is processed on the processing processor.