JPH04125738A - Data processor - Google Patents

Abstract

PURPOSE:To eliminate a need to generate and store a processing object data again by a management processor by sending a processing indication, which includes area information indicating the storage area of the processing object data in a data storage device and the data name, from the management processor to a processing processor. CONSTITUTION:If trouble occurs in the middle of processing in a processing processor 2, a management processor 1 generates a restart processing indication 20, which includes a start position START and an end position END of the processing object data group already stored in a magnetic disk device 3 and the data name included in a processing result report 30 received just before interruption, and transmits this indication 20 to the processing processor 2 through a communication line 4 when the processing is restarted. The management processor 1 terminates the processing when receiving the processing result report 30 for last data of the processing object data group from the processing processor 2.

Description

Detailed Description of the Invention [Field of Industrial Application] The present invention relates to a data processing device, and particularly to a data processing device having a configuration in which a management processor and a processing processor that can communicate with each other are connected to a common data storage device. Regarding.

[Prior Art] Conventionally, in this type of data processing device, a management processor stores a data group to be processed, including a plurality of pieces of data to be processed, in a data storage device such as a magnetic disk device, and creates an area indicating the storage area. The information is transmitted to the processing processor to instruct processing, and the processing processor performs predetermined processing such as collating all processing target data included in the storage area of the data storage device indicated by the received area information with data consisting of the data. , a processing form may be adopted in which a processing result with a data name included in each processing target data added is returned to the management processor.

By the way, when a processor fails during processing and restarts processing after recovering from the failure, it would be wasteful to restart the process from the beginning of the data group to be processed, so conventionally, It was like that.

For the processing target data that has been processed, the processing processor returns the processing result with the data name included therein to the management processor, so the data name added to the processing result that was returned immediately before the processing of the processing processor was interrupted. It is considered that processing was performed normally up to the processing target data with . Therefore, the management processor searches for a group of processing target data stored in the data storage device using the data name as a key, and selects the processing target data immediately after the found processing target data to the last processing target data as a new processing target. The data is stored again in the data storage device as a data group, and in the same way as normal processing, area information indicating the storage area is notified and the processing processor is made to perform processing. As a result, the processor processes the processing target data group included in the storage area indicated by the notified area information, that is, the processing target data that was not processed due to the interruption of processing, in the same way as normal processing.

[Problems to be Solved by the Invention] According to the conventional method described above, it is possible to avoid the waste of the processing processor reprocessing the processing target data that has been processed before the processing is interrupted, but for this purpose the management processor The process of re-creating the target data group and storing it in the data storage device must be performed. Note that the management processor searches the data storage device using the data name received just before the interruption, and retrieves area information indicating the storage area from the position of the data to be processed immediately after the data to be processed to the position of the last data to be processed. A method of notifying the processing processor and having it perform the processing is also considered, and it is thought that this will reduce the load on the management processor side to some extent. However, the remaining amount is small, and the number of parameters of area information may increase. In other words, normally, the beginning of the data group to be processed is stored at a physically well-separated point such as the beginning of the track of the magnetic disk device that is the data storage device, so the number of parameters for area information is small. , the position of the intermediate data to be processed is in the middle of the track, and parameters such as the number of hides from the beginning are required, and this cannot be used in cases where there is a limit to the number of parameters in the region information.

Therefore, an object of the present invention is to eliminate the need for a management processor to recreate or restorage of a data group to be processed, and to eliminate unprocessed data from a group of data to be processed stored in a data storage device due to processing interruption, etc. An object of the present invention is to provide a data processing device that allows a processor to process only a data portion to be processed using area information similar to area information transmitted before interruption.

[Means for Solving the Problems] In order to achieve the above object, the present invention provides a data processing device having a configuration in which a management processor and a processing processor that can communicate with each other are connected to a common data storage device. The management processor includes: a first means for sending a processing instruction including a data name and area information indicating a storage area of a data group to be processed stored in the data storage device to the processing processor; and a second means for sending a processing instruction that includes the data name but does not include the data name to the processing processor, and the processing processor receives the processing instruction that does not include the data name, thereby transmitting the received area information. third means for processing all of the data to be processed contained in the storage area of the data storage device indicated by and returning to the management processor a processing result with a data name included in each data to be processed; , by receiving a processing instruction including the data name,
From among the processing target data group stored in the storage area of the data storage device indicated by the received area information, the processing target data next to the processing target data containing the same data name as the processing target data to the last processing target data group and a fourth means for processing all of the processing target data up to the processing target data and returning a processing result with a data name included in each processing target data to the management processor.

Further, in a data processing device having another configuration, the second means and the third means are omitted, the management processor includes the first means, and the processing processor includes the fourth means. ing.

In the data processing device having this configuration, the management processor, when reprocessing after the processing of the processing processor is interrupted, adds the data name received from the processing processor immediately before the interruption to the same area information that was sent before the interruption. The added processing instruction is transmitted to the processing processor by the first means, and in cases other than restart processing after processing interruption, a data name included in the first processing target data of the processing target data group is added to the area information. The processing instruction is configured to transmit a processing instruction to the processing processor by the first means.

[Effect]

In the data processing device of the present invention including the first to fourth means, the management processor stores a processing target data group including a plurality of processing target data in the data storage device, and stores the processing target data group including the plurality of processing target data in the data storage device. When the second means sends a processing instruction including area information indicating a storage area to the processing processor, the third means of the processing processor receives the processing instruction and thereby processes the data storage device indicated by the received area information. It processes all of the processing target data included in the storage area, and returns the processing results to which the data names included in each processing target data are added to the management processor.

Furthermore, if a failure occurs in the processing processor during the above processing, upon resumption after recovery, the first means of the management processor receives the same area information from the processing processor immediately before the interruption that it sent before the interruption. By sending a processing instruction to which a data name has been added to the processing processor, and receiving a processing instruction including this data name, the fourth
The means selects the processing target data from the processing target data next to the processing target data containing the same data name as the data name among the processing target data group stored in the storage area of the data storage device indicated by the received area information. It processes all the processing target data up to the last processing target data in the group, and returns the processing results to which data names included in each processing target data are added to the management processor.

In the data processing apparatus including the first to fourth means described above, different types of processing instructions are used for normal processing and restart processing after processing interruption, but the present invention, which includes only the first and fourth means, In the data processing device, normal processing and restart processing after processing interruption can be performed only by processing instructions with data names added. That is, when restarting after processing is interrupted, operations similar to those of the data processing apparatus including the first to fourth means are performed, and during normal processing, the following operations are performed. First, the management processor uses the first means to transmit to the processing processor a processing instruction in which a data name included in the first processing target data of the processing target data group is added to area information. Then, the fourth means of the processing process 7 selects the processing target data containing the same data name as the data name among the processing target data group stored in the storage area of the data storage device indicated by the received area information. All processing target data from the next processing target data to the last processing target data of the processing target data group is processed, and the processing results with data names included in each processing target data are sent to the management processor. It performs the same operation as when restarting after suspending processing, which is to send a reply. Therefore, by setting the processing target data at the head of the processing target data group as data that does not need to be processed by the processing processor (dummy data), normal processing can be performed using the same procedure as restart processing after processing is interrupted. .

〔Example〕

Next, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a flowchart showing an example of processing by a processor included in a data processing device according to an embodiment of the present invention;
The figure is a flowchart showing an example of processing by a management processor included in the data processing device, and FIG. 3 is a block diagram of the data processing device. As shown in FIG. 3, the data processing apparatus of this embodiment includes a management processor 1 and a processing processor 2.
It includes a magnetic disk device 3 which is a data storage device commonly used by both processors 1 and 2. Further, both processors 1 and 2 can communicate with each other via a communication line 4.

FIG. 4 shows how the management processor 1 controls the magnetic disk device 3.
An example of the processing target data group stored in is shown below. here,
Data D1 to DN are processing target data that must be processed by the processing processor 2, and include names m1 to mN as data names in addition to variable length data to be processed.

FIG. 5 shows an example of the format of information exchanged between the management processor 1 and the processing processor 2 via the communication line 4. In FIG.

The processing instruction 10 and the restart processing instruction 20 are information sent from the management processor 1 to the processing processor 2. The processing instruction 10 includes a processing target data group start position 5TART and a processing target data group end position END, and the processing instruction 10 includes a processing target data group start position 5TART and a processing target data group end position END. Instructions 20 are the processing target data group start position 5TART, the processing target data group end position END, and the final processed data name LNA.
Including ME. Here, the processing target data group start position 5T
ART indicates the beginning position of the storage area of the magnetic disk device 3 in which the data group to be processed is stored, and END indicates the end position of the storage area, and the physical location of the magnetic disk device 3. This is what it points to. Further, the final processed data name LNAME is the data name included in the processing result notification received by the management processor 1 immediately before the processing processor 2 was interrupted.

The processing result notification 30 is information sent from the processing processor 2 to the management processor 1, and includes the processing result of the processing target data and the data name N included in the processing target data.
Including AME.

Next, the operation of this embodiment will be explained with reference to each figure.

Normally, the management processor 1 stores the data group to be processed as shown in FIG. 4 in the magnetic disk device 3 (501 in FIG. 2), creates the processing instruction 10 shown in FIG. 2 via communication line 4 (502
). Note that the data group to be processed is stored from a physically well-defined position such that its head coincides with the head of a track in the magnetic disk device 3, for example. after that,
The management processor 1 receives the processing result notification 30 shown in FIG. 5 from the processing processor 2 (503), and processes the data DN which is the final data of the processing target data group shown in FIG. When the result notification 30 is received (YES in 504), it is assumed that the processing of the current processing target data group by the processing processor 2 has been successfully completed, and the second
Finish processing the diagram.

On the other hand, when the processing instruction 10 shown in FIG. 5 is received from the management processor 1, the processing processor 2 sets itself to "data check N" which is a mode in which data names are not checked (601 in FIG. 1). Repeated position (RPTR)
Processing target data group start position 5 included in processing instruction 10
Set TART (603). Then, one piece of data is read from the location corresponding to the successive position (RPTR) of the magnetic disk device 3 (604). Since the mode is data check N (NO in 605), predetermined processing is performed on the read data. Execution (606) returns the processing result notification 30 shown in FIG. 5 including the data name of the processed data and the processing result of the data to the management processor 1 via the communication line 4 (607). Check whether the data at the included processing target data group end position END has been processed (610), and if it has not reached that position yet, update the successive position (RPTR) to the next data position (611). , and returns to step 604 to process the remaining data in the same manner as above. Further, when the data at the position of the processing target data group end instruction END has been processed, the processing shown in FIG. 1 is ended.

Next, an explanation will be given of the operation in the case where the processor 2 performing the above-mentioned processing encounters a failure during the processing and restarts the processing after recovery from the failure. In addition, the management processor l
It is assumed that the processing result notification 30 corresponding to the data Dn in FIG. 4 was last received from the processing processor 2.

When restarting the process, the management processor 1 performs the process 501 in FIG.
The processing target data group start position 5TART and the processing target data group end position END concerning the processing target data group already stored in the magnetic disk device 3, and the data name mn included in the processing result notification 30 received immediately before interruption. A restart processing instruction 20 shown in FIG. 5 including
5). Then, as in normal times, each time the processing result notification 30 shown in FIG. 5 is sent from the processing processor 2, it is received (503), and the final data of the processing target data group shown in FIG. When the processing result notification 30 for the DN is received (YES at 504), it is assumed that the processing by the processor 2 of the current data group to be processed has been successfully completed, and the processing shown in FIG. 2 is ended.

On the other hand, when the processing processor 2 receives the restart processing instruction 20 shown in FIG. 5 from the management processor 1, it is instructed to set itself to "data check Y" which is a mode for checking data names (step 602 in FIG. 1). The processing target data group start position 5TART included in the restart processing instruction 20 is set in the position (RPTR) (603). Then, one piece of data is read from the location corresponding to the read position (RPTR) of the magnetic disk device 3. Since the mode is 604L data check Y (YES in 605), the data name in the read data is the restart processing instruction. 30 final processed data names L
Check whether it is the same as NAME (60B). If they are not equal (NO in 608), restart processing instruction 2
Check whether the data at the end position END of the data group to be processed included in the processing target data group 610) has been processed, and if it has not reached that position yet, the read position
R) to the next data position, the process returns to 611L process 604 and the above-described process is repeated.

Further, in the process 608, if it is detected that the data name in the read data is the same as the final processed data name LNAM, E (YES in 608), the mode "Data check N" is selected in which the data name is not checked. ” 609) Proceed to process 610. Therefore,
Data read from this point on is subjected to the same processing as in normal times.

As described above, when restarting after interruption, the management processor 1
can proceed without re-creating or re-storing the data group to be processed.

Next, by intentionally providing dummy data that does not need to be processed at the beginning of a group of data to be processed, the present invention makes it possible to control normal processing and restart processing after interruption using only a processing instruction with a data name added. An example will be described. It should be noted that the configuration of the data processing device is the same as in the previous embodiment, as shown in FIG. 3.

FIG. 6 shows the management processor 1 controlling the magnetic disk device 3.
This is an example of the contents of a data group to be processed stored in , and dummy data DO having a known data name mO is added to the beginning of the data group to be processed in FIG. 4 already described.

FIG. 7 is a format example of a processing instruction sent from the management processor 1 to the processing processor 2, and the format is similar to the restart processing instruction 20 shown in FIG. However, in this embodiment, the processing instructions shown in FIG. 7 are used not only when restarting after processing interruption, but also during normal processing.
In FIG. 5, the part where the final processed data name LNAME is written is the data name PNAME. This data name PNAME is set to the data name mO of the data DO in FIG. 6 during normal processing, and when restarted after processing is interrupted, the last received data name is set as in the rabbit embodiment.

FIG. 8 shows an example of processing by the management processor 1, and FIG. 9 shows an example of processing by the processing processor 2. The difference between FIG. 8 and FIG. 2 is that in process 501', the data group to be processed as shown in FIG. The difference between FIG. 9 and FIG. 1 is that the formant processing instruction 40 shown in FIG. 7 is sometimes created, and the main difference between FIG. 9 and FIG. The point is that it has become.

In normal times, the management processor 1 stores a data group to be processed as shown in FIG. 6 in the magnetic disk device 3.
A processing instruction 40 in which the data name mO of the first data DO is set in E is transmitted to the processing processor 2. When the processing instruction 40 is received, the processing processor 2 performs the same operation as when receiving the restart processing instruction in FIG. 1. As a result, the process 606 is not performed on the first data DO, and instead, a match in the data name is detected in process 608, and the mode is changed to "Data Check NJ" in process 609. Therefore, Subsequent data D1 to DN
is processed normally.

Further, when restarting after processing interruption, the management processor 1 issues a processing instruction 40 in which the data name PNAME shown in FIG. 7 is set to the data name included in the processing result notification 30 shown in FIG. Create and process processor 2
(505' in FIG. 8). When the processing instruction 40 is received, the processing processor 2 performs the processing shown in FIG. This processing is the same as the normal processing described above and the processing when the restart processing instruction is received in FIG. In this way, according to this embodiment, the processor 2 does not need to perform separate processes during normal times and when restarting, and the program for the processor 2 can be simplified.

〔Effect of the invention〕

As explained above, according to the present invention, the following effects can be obtained.

(1) When the processing processor resumes processing after it has been interrupted, the management processor as in the past recreates the data group to be processed;
There is no need to restorage, and the amount of processing by the management processor can be reduced.

(2) In the processing processor as well, it is only necessary to perform substantial processing on only the part of the data to be processed that is unprocessed due to processing interruption among the data group to be processed stored in the data storage device, and the amount of processing is lower than that of the previous one. There is almost no difference.

(3) Processing instructions can be given to the processor using the same area information as the area information used in normal times.

(4) According to a data processing device that includes the first and fourth means and is configured to provide data that does not need to be processed by a processor at the beginning of a group of processing target data stored in a data storage device, one type of processing instruction is provided. It is possible to control normal processing and restart processing after processing interruption, and in particular, the program on the processor side can be simplified.

[Brief explanation of drawings]

FIG. 1 is a flowchart showing an example of processing by a processing processor in an embodiment of the present invention; FIG. 2 is a flowchart showing an example of processing by a management processor in an embodiment of the invention; FIG. 3 is a flowchart showing an example of processing by a management processor in an embodiment of the invention. 4 is a diagram showing a configuration example of a data group to be processed in an embodiment of the present invention. FIG. 5 is a diagram showing a configuration example of a data group to be processed in an embodiment of the present invention. FIG. 6 is a diagram showing an example of the structure of a data group to be processed in another embodiment of the present invention. FIG. 7 is a diagram showing an example of the content of information to be exchanged. FIG. FIG. 8 is a flowchart showing an example of processing by a management processor in another embodiment of the present invention; FIG. 9 is a flow chart showing an example of processing by a management processor in another embodiment of the present invention. 2 is a flowchart showing an example of processing. In the figure, l...Management processor 2...Processing processor 3...Magnetic disk device 4...Communication line 10.40...Processing instruction 20...Resuming processing instruction 30...Processing result notification Processing example of the management processor Fig. Configuration example of the data processing device of the present invention Fig. An example of a group of data to be processed Fig. 10 Processing instruction TART ND AME Fig. Example of content of information exchanged between the management processor and the processing processor Fig. Object to be processed Another example of a data group Figure: Example of the contents of a processing instruction sent from the management processor to the processing processor Figure: Another example of processing by the management processor Figure: Another example of processing by the processing processor

Claims (3)

[Claims] (1) In a data processing device having a configuration in which a management processor and a processing processor that can communicate with each other are connected to a common data storage device, the management processor stores a group of data to be processed stored in the data storage device. a first means for sending a processing instruction including area information indicating an area and a data name to the processing processor; and a first means for sending a processing instruction including the area information but not the data name to the processing processor. 2, wherein the processing processor receives the processing instruction that does not include the data name, and processes all the processing target data included in the storage area of the data storage device indicated by the received area information. a third means for returning a processing result obtained by performing processing and adding a data name included in each processing target data to the management processor; and receiving a processing instruction including the data name;
From among the processing target data group stored in the storage area of the data storage device indicated by the received area information, the processing target data next to the processing target data containing the same data name as the processing target data to the last processing target data group and a fourth means for processing all the processing target data up to the processing target data and returning a processing result with a data name included in each processing target data to the management processor. data processing equipment. (2) In a data processing device having a configuration in which a management processor and a processing processor that can communicate with each other are connected to a common data storage device, the management processor stores a group of data to be processed stored in the data storage device. comprising a first means for sending a processing instruction including area information indicating an area and a data name to the processing processor, and the processing processor: By receiving the processing instruction including the data name,
From among the processing target data group stored in the storage area of the data storage device indicated by the received area information, the processing target data next to the processing target data containing the same data name as the processing target data to the last processing target data group It is characterized by comprising a fourth means for processing all the processing target data up to the processing target data and returning a processing result with a data name included in each processing target data to the management processor. Data processing equipment. (3) When reprocessing after the processing of the processing processor is interrupted, the management processor sends a processing instruction to the first In cases other than restart processing after processing interruption, the first means sends a processing instruction to the processing processor by adding a data name included in the first processing target data of a processing target data group to the area information. 3. The data processing apparatus according to claim 2, further comprising a configuration for transmitting the data to the processing processor by.