JPH04107034A - Retry frequency learning device - Google Patents

Abstract

PURPOSE:To prevent the degradation of the communication efficiency accompanied with system expansion or the like by automatically correcting the frequency in retry for the occurrence of communication error in each external device in accordance with environments of the system. CONSTITUTION:Plural POS terminals 1 consisting of electronic cash registers and one POS controller 2 consisting of a host computer are connected through a transmission line 3 like an BDLC line to constitute the system. The POS controller 2 sets required data stored on a set data file 4 to each POS terminal 1 through the transmission line 3 and gathers sales data of goods sold and registered by each POS terminal 1 to totalize data for each transaction classification in a sales totalization file 5 and analyzes sales in the whole of a shop.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a retry count learning device suitable for use in a network system such as a PO5 (point of sale) system.

C. Prior Art] In general, a POS system connects multiple POS terminals consisting of electronic cash registers and one host controller consisting of a host computer through a transmission line such as an HDLC (High Level Data Link Φ Control) line. It is configured by connecting through. Then, the host controller sets the necessary data for each POS terminal via the above transmission path, collects sales data of products registered for sale at each POS terminal, aggregates it by transaction, etc. It is designed to perform overall sales analysis, etc.

By the way, in a network system such as a POS system in which multiple terminals (upper controller and POS terminal) with communication functions are connected via a transmission path, one terminal communicates with the other station. If so, first try to establish a line connection with the other station (connection establishment task). Next, the necessary data is transmitted through the connected line (data transfer operation). After that, the line is disconnected (connection disconnection operation). In this case, if a communication error occurs due to no response or an abnormal response from the partner station when each task is executed, the same task is retried (re-executed) a preset number of times. If the error condition is not resolved even after retrying a preset number of times, the current communication process is terminated as an error.

Conventionally, the number of retries when a communication error occurs is given to each terminal as a parameter when the system is constructed, and it remains fixed even if the system is changed. However, when the system is expanded, the probability that a communication error will occur tends to increase, and conversely, when the system is downsized, the probability that a communication error occurs tends to decrease. For this reason, it can be said that it is preferable from the viewpoint of communication efficiency to set the number of retries suitable for the system environment.

[Problem to be solved by the invention] However, in the past, the number of retries when a communication error occurred was given to each terminal as a parameter and fixed when the system was constructed, so communication efficiency decreased due to system expansion, etc. was there.

Therefore, the present invention provides a retry count learning device that can automatically modify the retry count when a communication error occurs in each external device according to the system environment, and can prevent a decrease in communication efficiency due to system expansion. This is what we are trying to provide.

[Means for Solving the Problems] The present invention provides error information collection means for collecting error information generated in communication processing between external devices connected via a transmission path, retry count determining means for automatically determining the retry count for an error occurring during communication processing in the external device based on error information; and retry count setting means for setting the retry count determined by the determining means in the external device. This is a retry number learning device equipped with the following.

[Operation] In the present invention configured in this manner, error information generated by communication processing of an external device is periodically collected, and the number of retries is determined based on the collected error information. For example, if the system is expanded and communication errors increase, the current number of retries will be increased; conversely, if the system is downsized and communication errors decrease, the current number of retries will be decreased. Correct the direction and decide. Then, the determined number of retries is set in the external device. As a result, from now on, the external device will retry the latest number of retries every time a communication error occurs.

[Embodiment] Hereinafter, an embodiment in which the present invention is applied to a POS system will be described with reference to the drawings.

FIG. 2 is a conceptual diagram of a POS system, in which multiple POS terminals 1 consisting of electronic cash registers and one POS controller 2 consisting of a host computer are connected via a transmission line 3 such as an HDLC line. It is composed of Then, the POS controller 2
The necessary data stored in the setting data file 4 is set for the terminal 1 via the transmission path 3, and the sales data of products registered for sale at each POS terminal 1 is collected and stored in the sales summary file 5. It is now possible to aggregate sales by transaction and perform sales analysis for the store as a whole.

FIG. 3 is a control block diagram of the POS controller 2. As shown in FIG. That is, a ROM (read-only memory) 13 in which communication programs and the like are stored is connected to a CPU (central processing unit) 11 as a main body of the control unit via a pass line 12;
RAM (memory that can be read and written at any time) 14 in which a work buffer 7 used for data processing is formed, and E2FROM (electrically erasable programmable ROM)
a transmission control circuit 16 for controlling data communication between each of the storage units 15 and each PO8 terminal 1 via the transmission line 3, a disk device in which the setting data file 4 and the sales summary file 5 are provided; 17, a disk driver 18, a full keyboard 19. It is constructed by connecting an input/output device interface 22 to which external input/output devices such as a CRT display 20 and a printer 21 are connected, and each part of a clock circuit 23 that measures date and time.

FIG. 4 is a control block diagram of the POS terminal 1. That is, "registration", "payment", "setting", etc.
A ROM 33 that stores programs and the like for executing various business modes as an OS terminal. RAM3 in which memory etc. for registering and processing product sales data is formed.
4 and E2 A transmission control circuit 36 that controls data communication between each memory section of the PROM 35 and the POS controller 2 via the transmission line 3, and a signal from the mode switch 37 for selecting the various business modes. I10 outputs an open signal to the drawer 38 at the same time as input
A boat 39, a keyboard 40 for inputting information regarding the products for sale, a display 41 for displaying the price of the products for sale, the customer's total amount, etc., a printer 4 for printing on receipts, etc.
2 is connected to the input/output device interface 43, and other parts.

Here, a communication management table 50 shown in FIG. 1 is formed in the E2 PROMs 15 and 35 of the POS controller 2 and each POS terminal 1, respectively. This communication management table 50 includes an area for storing the total number of execution requests for each of the above-mentioned "connection establishment", "data transfer" and "connection disconnection" operations that occur during data communication with the other station. A1, area A2 for storing the total number of retry occurrences in the relevant business;
There is also provided an area A3 for storing the set value of the number of retries for the corresponding task.

Further, a retry counter 60 shown in FIG. 5 is formed in the RAMs 14 and 34 of the POS controller 2 and each POS terminal 1, respectively.

Furthermore, in the ROM 13 of the POS controller 2,
A learning table 70 shown in FIG. 6 is stored. In this learning table 70, an appropriate retry count setting value is set in advance in accordance with the ratio of the number of retries to the number of communication service requests.

Therefore, the POS controller 2 and each PO39-
The CPU II and 31 of Mi+tll are programmed to execute the control shown in FIG. 7 when a communication service occurs with a partner station during system operation. That is, when a request for execution of any of the tasks "establishment of connection,""transfer of data," and "disconnection of connection" occurs, the E2 PROM corresponds to the corresponding task item in the communication management table 50 formed in its own E2 PROM 15 and 35. Add "+1" to the number of requests generated in area A1.

Further, the retry count setting value of the area A3 corresponding to the corresponding work item in the communication management table 50 is set in the retry counters 60 of the RAMs 14 and 34. After that,
Communication processing for the relevant business is executed according to a predetermined communication procedure. If the communication process is completed without any communication error occurring, the control proceeds to the next step.

On the other hand, if a communication error occurs because the other station does not respond or an abnormal response is returned during communication processing, the retry counter 60 is counted down by "-1". Then, if the count value is "0", add "+1" to the number of retries in area A2 corresponding to the corresponding business item in the communication management table 50, and
Execute the communication process for the relevant task again.

In this way, each time a communication error occurs, the retry counter 60 is counted down by "-1", and when the count value reaches 0[deg.], the current communication process is terminated as an error.

Therefore, area Al of the communication management table 50 formed in the E2 PROMs 15 and 35 of the POS controller 2 and each POS terminal 1.

A2 stores the total number of execution requests for the corresponding communication service and the total number of retries due to communication errors. The stored value is retained even if the system is powered down.

Further, the CPU II of the POS controller 2 is programmed to execute the control of the retry count learning function shown in FIG. 8 in the initial processing at the time of system start-up. That is, first, communication management information is collected for each POS terminal 1 connected via the transmission line 3. As a result, each POS terminal 1 sends the number of request occurrences and the number of retry occurrences for each item held in the communication management table 50 of the E2FROM 35, which is collected, and the collected data and its own E
The relevant data held in the communication management table 50 of the 2FROM 15 is aggregated for each item. (Error Information Collection Means) Next, the learning process shown in FIG. 8 is executed for each item to determine the retry count setting value for each item. First, the total number of request occurrences is compared with a preset reference value N (≧2). If the number of request occurrences is less than the standard value, there will be a shortage of samples and they will not be useful as learning data.
The current retry count setting value set in area A3 of the communication management table 50 is not changed, and the contents of area A3 are determined as the retry count setting value of the corresponding item.

On the other hand, if the number of requests generated is equal to or greater than the reference value N, the ratio of the number of retries generated to the number of requests generated is calculated. Then, the learning table 70 stored in the ROM 13 is searched using the calculated ratio value, and the corresponding retry count setting value is read out and determined to that value. (Retry count determining means) Once the retry count set value is determined by executing the above learning process for each item, first the area of the communication management table 50 formed in the own E2 PROM 15 is determined according to the determined value. Amend the contents of Section 8.3. Next, each POS terminal 1 is connected via the transmission line 3.
A request is made to modify the retry count setting value. This results in
Since the contents of area A3 of the communication management table 50 formed in the E2 PROM 35 of each POS terminal 1 are automatically corrected, this process is ended after confirming that the correction has been made. (Retry count setting means) As described above, in this embodiment, the POS that functions as an external device
In the controller 2 and each POS terminal 1, the number of retries is managed by a communication management table 50 as information on errors that occur during communication processing with the other station. Then, each time the system is started up, the POS controller 2 collects and aggregates information on communication errors that have occurred in the system, and automatically determines the retry count setting value based on the aggregated information. The determined retry count setting value is set in each POS terminal and the POS controller 1 itself. Specifically, if the number of times communication errors occur is large, the ratio of the number of retries to the number of requests will be large, so the retry count setting value is set to a large value; conversely, if the number of communication errors is small, the number of retries is set to a large value Since the ratio of the number of occurrences becomes small, the retry count setting value is set small. Therefore, during operation after system startup, if an error occurs in the POS controller 2 and each PoS terminal 1 during communication processing with the other station, retries will be performed for the number of retries set at system startup. Becomes exposed.

Generally, in a POS system, if you try to expand the system by increasing the number of POS terminals 1, the probability of a communication error occurring tends to increase; conversely, if you try to downsize the system by decreasing the number of POS terminals 1. There is a tendency for the probability of communication errors to occur to be lower. Therefore, when the system is expanded and the number of communication errors tends to increase, the number of retries increases, increasing the possibility that communication processing will be performed normally. As a result, the same level of communication efficiency as before the system expansion can be obtained, and the system can be expanded easily.

Note that the present invention is not limited to POS systems, but can be applied to general network systems in which external devices are connected to each other via transmission paths. Further, in the embodiment described above, the number of retries is managed as error information in communication processing, but the number of retries is not limited to this, and the number of retries may be learned by taking into account other factors.

Further, in the embodiment described above, an example was shown in which learning the number of retries is performed using the learning table 70, but it may be performed by calculation using a calculation formula.

It goes without saying that various other modifications can be made without departing from the gist of the present invention.

[Effects of the Invention] As described in detail above, according to the present invention, the number of retries when a communication error occurs in each external device can be automatically corrected in accordance with the system environment, which is useful for system expansion, etc. It is possible to provide a retry count learning device that can prevent the accompanying decrease in communication efficiency.

[Brief explanation of drawings]

The figures are diagrams showing an embodiment in which the present invention is applied to a POS system, in which Fig. 1 shows a communication management table, and Fig. 2
The figure is a conceptual diagram of the POS system, Figure 3 is a control block diagram of the POS controller, Figure 4 is a control block diagram of the POS terminal, Figure 5 is a diagram showing a retry counter, and Figure 6 is a diagram showing a learning table. , FIG. 7 is a flowchart showing the control routine for communication operations in the POS controller and POS terminal, FIG. 8 is a flowchart showing the control routine for the retry count learning task in the POS controller, and FIG. 9 is the control routine for the retry count learning task. 3 is a flowchart showing a retry setting value learning process in FIG. 1... POS terminal, 2... POS controller, 3... Transmission line, 11.
31...CPU, 16.36...Transmission control m circuit, 5
0...Communication management table, 60...Retry counter, 70...Learning table. Applicant's Representative Patent Attorney Takehiko Suzue Figure 2 Figure 6 Figure Figure

Claims (1)

[Scope of Claims] Error information collection means for collecting error information generated in communication processing between external devices connected via a transmission path; The apparatus further comprises a retry count determining means for automatically determining the retry count in response to an error occurring during communication processing in the device, and a retry count setting means for setting the retry count determined by the determining means in the external device. A retry number learning device characterized by: