JPS6243761A - Working state monitor system for on-line message processing system - Google Patents

Abstract

PURPOSE:To detect the working defect of an on-line message processing system by using an analyzing means for the working state of a message processing system and a warning means which delivers the warning information if an over-working state is decided with the processing system. CONSTITUTION:When a load factor L(t1) is obtained at the next acquiring point t1, a warning means 15 checks whether the factor L(t1) exceeds or not a prescribed level. If the factor L(t1) exceeds the prescribed level, a wrong working state is decided. Then a warning message is displayed on a console 44 together with the number of inputs, the queuing number and the execution end number counted at four time points, i.e., the point t1 and its three preceding points t-1-t-3. Here it can be easily estimated that such control is possible to stop temporarily the acceptance of the message at a message processing part A until the wrong processing is deleted by delivering an alarm to the part A. Thus it is possible to take a measure to a working defect or to study its factors at an early stage.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a system for monitoring the operating status of an online message processing system.

(Conventional techniques) Conventional online message processing systems: - Receive messages input from a terminal, temporarily store them in a message queuing area, and then extract messages one by one from the message queuing area using a message sorting means. A series of processes such as sending the message to the message execution control means, processing according to the message content from the message execution control means G, and outputting a response message to the terminal are realized by executing the message processing program. There was no means to monitor the operating status of the message processing system.

[Problem that the invention seeks to solve]

By the way, in such systems, it is of course important to ensure online performance, but due to incorrect programming of the online message processing system or exclusive control of the database, processing waiting time may become so long that the online performance is compromised. , or the process sometimes stops progressing. In such a case, in the 9th period, the lie will be detected, the malfunction will be discovered, and the cause will be investigated.
It is desirable to immediately collect data necessary for improvement and to take measures against malfunctions at an early stage.

However, as mentioned above, conventional online message processing systems do not monitor the operating status during operation, and it is not until the abnormality is recognized by the person operating the terminal that the operating status is abnormal. Anomalies are only detected when the problem is recognized, or when analysis is performed periodically every hour or two hours. Therefore, data collection for investigating the cause or making improvements is delayed, and malfunctions are often detected. The drawback was that it was not possible to take countermeasures or investigate the cause at an early stage.

SUMMARY OF THE INVENTION An object of the present invention is to provide an operating state monitoring method for an online message processing/system that can eliminate the above drawbacks and immediately detect malfunctions in the online message processing system.

(Means for Solving the Problems) In order to achieve the above object, the present invention includes: a message receiving stage for receiving a message input from a terminal and storing it in a message queuing area; Message distribution means for retrieving messages; message execution control means for performing processing according to the message retrieved by the cough message distribution means; and message output for outputting a response message to a terminal upon completion of execution of the message control rod control means. In an online message processing system having means, an input number acquisition means for acquiring the number of messages input from a terminal, and a queuing number for acquiring the number of messages that have been input but are waiting to be processed. an acquisition means; an execution completion number acquisition means for acquiring the number of execution completions of messages processed by the message execution control means; and data acquired by the input number acquisition means, the queuing number acquisition means, and the execution completion number acquisition means. A data storage means for storing data in a data storage area, a data reading means for reading data stored in the data storage area, and an operating state of the message processing system is analyzed based on the data read by the data reading means. An analysis means and a warning means for determining the operating state analyzed by the analyzing means and outputting warning information if the operating state is equal to or higher than a predetermined operating state are provided.

(Operation) The operating state of the message processing system is the number of inputs of the data storage means read by the data reading means.

This is analyzed by an analysis means based on the number of queues and the number of completed executions, and if the processing wait time becomes so long that it impairs online performance due to program fraud, or if the processing hardly progresses, the operating state will change to the specified operating state. The warning means indicates that the patient's condition has become 9 or higher on the ipsilateral side, and a warning is issued.

[Embodiment] Figure 1 is a block diagram of an embodiment of the present invention, in which a terminal l
Message receiving means 2 receives (ε) an input message from a message queuing area 6 and stores it in a message queuing area 6; a message sorting means 3 extracts a message from the message queuing area 6; and a message execution system that performs processing according to the content of the message. '+B means 4 and message output means 5 which returns a response message to the terminal 1 upon completion of execution. As shown in FIG.
A message number storage area 7 is provided to store the number of messages that have been input but are waiting to be processed (queuing number) and the number of messages that have been executed (execution completion number), and input number acquisition means 8 , keying number acquisition means 9
．． Obtain number of completed executions 10 stages 10. Data storage means 11. Data storage area 12. Data reading means 13. Analysis means 14
and a warning means 15, and an operating state monitoring section B that operates independently of the online message processing section A.
It has been established.

In the first M, when a message is input from the terminal 1, the message receiving means 2 receives the input message by processing as shown in FIG. 3 (516),
Next, 1 is added to the input number in the message number storage area 7 shown in FIG. 2 (S17). Also, store the received message in the message queuing area 6 (318).
, 1 is added to the queuing number in the message number storage area 7 (S19).

The message distribution means 3 performs the process shown in FIG. 4, for example, while the message execution control means 4 is not executing a transaction, and if there is a keyed message in the message queuing area 6. For example, one message is extracted from there and sent to the message execution control means 4 (520). 521), calculates the queuing number in the message number storage area 7 (S
22).

When a message is sent from the message distribution means 3 to the message execution fi112B means 4, the message execution control means 4 executes a transaction corresponding to the sent message.

When the message execution system tIm means 4 completes the execution of the message, the message output means 5 receives it and adds one to the number of completed executions in the message number storage area 7, as shown in FIG. 5 (323), and executes the message. Notify message distribution means 3 of completion (524), terminal l
A response message is returned to (S25).

On the other hand, the number of inputs, the number of queuing, and the number of completed executions stored in the message number storage area 7 are obtained by the input number acquisition means 8.
．． Queuing number acquisition means 9. Execution completion number acquisition means 10
, and the data storage means 11 stores in the data storage area 12 the current number of queues, the number of messages inputted between the current time and a predetermined time before (the number of inputs), and the number of completed executions. Ru. This is done as follows.

For example, as shown in FIG. 6, the data storage means 11 notifies the timer device P, 34 in step S33, receives a notification from the timer function 34 after a certain period of time has elapsed in step 335, and then performs steps 336 to 338. Execute.

That is, steps 336 to 338 are executed at predetermined time intervals. Here, the time point at which the notification is received in step S35 is defined as the current acquisition point t0, the time point at which the previous notification from the timer function 34 is reported is defined as the previous acquisition point 1-, and the time point at which the next notification after to is notified is defined as the degree score t1.

When the data storage means 11 receives the notification from the timer, the input number acquisition means 8. The queuing number acquisition means 9 and the execution completion number acquisition means 10 are started, and each means is activated, for example, by the seventh
The input number acquisition means 8 performs the processing shown in FIGS. 526) and memorize its value for calculating the degree score t1 527), the number of inputs at the current acquisition time t0 and the previous acquisition time 1-. The difference between the number of inputs and the number of inputs is determined as the number of inputs in that section (328). Also, queuing number acquisition means 9
Now, as shown in FIG. 8, the queuing number at the current acquisition point t0 is obtained by reading the queuing number in the message number storage area 7 (529), and the execution completion number obtaining means 10
Now, as shown in FIG. 9, by reading the number of completed executions in the message number storage area 7, we can obtain the number of completed executions up to the current acquisition point t0 (530), and use that value for the calculation in the next acquisition panic [1]. Please remember 531), the current acquired point t0 and the previous acquired point 1-,
The difference in the number of completed executions in that section is determined as the number of completed executions in that section (S32). The data storage means 11 includes an input number acquisition means 8, a queuing number acquisition means 9. Execution completion number acquisition means l
At O, the previous acquired point 1- and the current acquired point t are obtained as described above.

The number of inputs, number of completed executions, and current acquired points in the interval with [
. Once the queuing number is determined, it is acquired (336) and its value is stored in the data storage area 12 (
S37). This operation of the data storage means 11 is performed at predetermined time intervals until an instruction to terminate the monitoring operation is given.The data storage area 12 stores, for example, the number of inputs within a certain time interval, the number of completed executions, and the number of completed executions as shown in FIG. The queuing numbers at the time these are calculated are stored for each section.

Now, the data reading means 13 reads the current acquisition point every time new data is written into the data storage area 12 by the data storage means 11. As shown in Fig. 11, the current acquired point [. and【. 3 acquired points t-1, t
-2, t-3, the number of inputs, the number of queuing, and the number of completed executions for a total of four acquisition points are read from the data storage area 12 (
S39), the analysis means 14 performs processing as shown in FIG. 11 based on this read value to calculate the order score [
The load factor L (Ll) at 1 is calculated. This load factor L
(tl) is calculated, for example, as follows.

The analysis means 14 obtains 0°1 from the data reading means 13,
．． 1t, 1. The number of inputs at each acquisition point is 1(b)
(i=o, 1.-2. 3), the cubic equation % equation % which shows the relationship between the time t and the number of inputs 1(t) assuming that the line connecting each point is linear. However, a, , a Yo+a3. a4 is a coefficient), and the number of inputs 1 (tl) at the order score [l is calculated from this equation. Similarly, the number of completed executions is E(t,)(i=0.-1.
-2. -3), and the cubic formula % formula that shows the relationship between the time t and the number of completed executions E(t). The execution completion viE (tl) at t1 is calculated (S40).

Then, the number of queuing at the current acquisition point t0 is Q(to)
Then, using the values of Q(to), E(tl), and +(tl), the ratio of the number of queuing to the number of completed executions at the degree score t is calculated as the load factor L (tl) using the following formula. (S41).

L(t,)=(Q(to)+l(L+))/E
(tl) Figure 13 shows the 3 obtained by the analysis means 14.
This is a diagrammatic representation of the following equations 1(t) and ED), and the temporal change in the number of queuing is shown as Q (t).

When the load factor L(1+) of the order score t1 is determined in this way, the warning means 15 displays the load factor 1.0 as shown in FIG. 14, for example. (1+) is a preset value, for example 1.5
342), and if it does, the operating state is invalid and a warning message is sent, along with the current acquired point t0 and the previous 3 acquired points 1-, 1, stored in the data storage area 12. -1.1-, the number of inputs, the number of keyings, and the number of completed executions at a total of four points in time are displayed on, for example, the console 44 (S43). In this case, by issuing a warning to the message processing section A, it is easy to have the message processing section perform controls such as temporarily stopping accepting messages until the improper processing can be eliminated. I can guess.

In this way, in this embodiment, even if the operating state of the online system processing unit is not abnormal at present, if it is estimated that it will become abnormal at the next acquisition point, a warning is issued, so malfunctions can be detected more quickly. It becomes possible to discover. In addition, as a method for monitoring the operating status online, it is possible to incorporate an operating status monitoring program into the online message processing program, that is, to perform message processing and monitoring processing in the same space. However, in this embodiment, the message processing Since the monitoring processing is placed in a space different from the operating space and is configured so that it can operate independently, the monitoring processing does not interfere with the message processing and impair online performance.

Although one embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and various other additions and changes are possible. For example, the addition rate L(L.) of the current acquisition is expressed as L(L.)-(Q(t-,) + l(to))/
E(t6) may be used as a parameter indicating the operating state, or other parameters such as the queuing number itself may be used as a parameter indicating the operating state. That is, in the present invention, any parameter can be used as long as it changes depending on the operating state.

〔Effect of the invention〕

As explained above, according to the present invention, if the operating state of the online message processing system becomes abnormal, a warning is immediately issued, so that processing can be delayed due to malfunction in the program of the online message processing system or exclusive control of data storage. It is possible to immediately detect that the online time has decreased to 11 or that the processing has hardly progressed. Therefore, when detecting a malfunction, you can promptly investigate the cause or collect data for improvement. This has the effect that countermeasures for malfunctions and investigation of causes can be taken at an early stage. Furthermore, when a warning is issued, an emergency zF device can be used as needed, such as inputting more urgent messages from the terminal with priority over less urgent messages.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2 is an explanatory diagram of the contents of the message number storage area 7, FIG. 3 is an explanatory diagram of the operation of the message receiving means 2, and FIG. 4 is an explanatory diagram of the message distribution means 3. 5 is an explanatory diagram of the operation of the mesomo-2 output means 5, FIG. 6 is an explanatory diagram of the operation of the data storage means 11, FIG. 7 is an explanatory diagram of the operation of the input number acquisition means 8, and FIG. 8: FIG. 9 is an explanatory diagram of the operation of the queuing number acquisition means 9; FIG. 9 is an explanatory diagram of the operation of the execution completion number acquisition means 10;
11 is an explanatory diagram of the contents of the data storage means 11, FIG. 11 is an explanatory diagram of the operation of the data reading means 13, and FIG. 12 is an illustration of the analysis means 14.
FIG. 13 is an explanatory diagram of the method of calculating the negative part rate, and FIG. 14 is an explanatory diagram of the operation of the warning means 15. In the figure, ■ is a terminal, 2 is a message receiving means, 3 is a message distribution means, 4 is a message execution control means, 5 is a message output means, 6 is a message queuing area, 7 is a message number storage area, and 8 is the number of inputs 9 is a queuing number obtaining means, IO is an execution completion number obtaining means, 11 is a data storage means, 12 is a data storage area, 13 is a data reading means, 14 is an analysis means, 15 is a warning means, A is online The message processing section B is an operating state monitoring section. Figure 2 is an explanatory diagram of the contents of Metsu Seven-Ji Raw Number Memory Castle 7. Figure 3 is an explanatory diagram of the operation.
FIG. 4 is an explanatory diagram of the operation of the message output means 5. FIG. 5 is an explanatory diagram of the operation of the data storage means 1. FIG. 6 is an explanatory diagram of the operation of FIG. Figure 10 Diagram explaining the operation of the analysis means 14 Figure 12

Claims (1)

[Scope of Claims] Message receiving means for receiving a message input from a terminal and storing it in a message queuing area; Message sorting means for taking out a message from the message queuing area; and a message taken out by the message sorting means. In an online message processing system, the online message processing system includes a message execution control means that performs processing according to the message execution control means, and a message output means that outputs a response message to a terminal upon completion of execution of the message execution control means. an input number acquisition means for acquiring the number of messages; a queuing number acquisition means for acquiring the number of queued messages that have been input but are waiting to be processed; and a number of completed executions of messages processed by the message execution control means. Execution completion number acquisition means for acquiring the number of executions; data storage means for storing data acquired by the input number acquisition means, the queuing number acquisition means, and the execution completion number acquisition means in a data storage area; data reading means for reading the data read by the data reading means; analysis means for analyzing the operating state of the message processing system based on the data read by the data reading means; 1. An operating state monitoring method for an online message processing system, comprising: a warning means for outputting warning information if the operating state is higher than the operating state.