WO2016203543A1

WO2016203543A1 - Data collection device and method

Info

Publication number: WO2016203543A1
Application number: PCT/JP2015/067292
Authority: WO
Inventors: 篠原　剛; 英嘉坂平; 哲男門脇
Original assignee: 株式会社日立製作所
Priority date: 2015-06-16
Filing date: 2015-06-16
Publication date: 2016-12-22

Abstract

[Problem] To propose a highly reliable data collection device and method. [Solution] A data collection device for collecting and accumulating a plurality of types of data transmitted from one or a plurality of terminals, and a data collection method performed by the data collection device, wherein: the priority currently assigned to each data type is determined on the basis of past search and delivery requests, the number of uncollected sets of data of each data type is obtained, and the current data processing capacity of the data collection device is detected; and, after recovery from a failure, a data collection scheme for collecting uncollected data is determined taking into account the priority currently assigned to each data type, the number of uncollected sets of data of each data type, obtained after the recovery from the failure, and the current data processing capacity of the data collection device, detected after the recovery from the failure, and then the uncollected data are collected in accordance with the determined collection scheme.

Description

Data collection apparatus and method

The present invention relates to a data collection device and method, for example, a data collection and distribution system that accumulates data of a plurality of data types transmitted from a plurality of data collection source terminals and distributes the accumulated data in response to a request from an external device. It is suitable for application to.

In recent years, interest in the use of big data has increased. Among the data that is assumed as big data, in addition to the conventional information system data such as POS (Point Of Sales) system and transaction data, it is output from sensor devices mounted on construction machines, turbines, etc. Control system data is also included.

As a system for utilizing such big data, there is a server device (hereinafter referred to as a data collection and distribution server) that collects data from a large number of terminals (hereinafter referred to as data collection source terminals). In response to a request from a service user, a data collection and distribution system that provides the service user with the requested data from the data collected by the data collection and distribution server has been put into practical use.

Note that Patent Document 1 discloses a technique for transmitting data in order according to priority when data is transmitted between computers.

Japanese Patent Laid-Open No. 11-194978

By the way, in such a data collection and delivery system, the data collection and delivery server is required to deliver the data required by the service user as quickly as possible even when an irregular situation occurs. This is because if the delivery of data requested by the service user is delayed, the service user's business delay occurs, which may adversely affect the service user's business.

Therefore, in such a data collection / distribution system, for example, when a failure that cannot collect data from each data collection source terminal occurs, there is a high possibility of being requested by a service user after recovery from the failure ( If the data collection and distribution server can be given a function that collects data from the data collection source terminal more preferentially (higher priority as viewed from the service user), it will have a negative effect on the service user due to such failure. It is considered that the reliability of the data collection and distribution server can be improved by reducing the amount.

In addition, since the data collection and distribution server immediately after failure recovery is operated in a degenerate state with reduced data processing capacity, when providing this function to the data collection and distribution server, data should be flowed at a flow rate that matches the data processing capacity of the data collection and distribution server. A mechanism to collect the data is also required.

The present invention aims to solve the above problems all at once, and proposes a data collection and distribution apparatus and method capable of improving reliability.

In order to solve such a problem, in the present invention, data of a plurality of data types transmitted from one or a plurality of terminals is stored, and the stored data is dealt with in response to a search distribution request from an external device. In a data collection device that distributes the data of the data type to the external device, a priority detection unit that detects the priority of each of the current data types based on the past search distribution request, and for each data type A data number acquisition unit that acquires the number of data of the uncollected data, a server state detection unit that detects the current data processing capability of the device, and each of the current data types detected by the priority detection unit Priority, the number of data acquired by the data count acquisition unit after recovery from the failure, and the server status detection unit after recovery from the failure A data collection control unit for deriving a collection method of the uncollected data in consideration of the current data processing capability of the detected own device, and collecting the uncollected data according to the derived collection method; I made it.

Further, in the present invention, data of a plurality of data types transmitted from one or a plurality of terminals is accumulated, and data corresponding to the data type is stored from the accumulated data in response to a search distribution request from an external device. In the data collection method executed by the data collection device that distributes the data to the external device, the first step of detecting the priority of each current data type based on the past search distribution request, and the data Acquire the number of data of the uncollected data for each type, and the second step of detecting the current data processing capability of the own device, the current priority of each data type, and after recovery from a failure Accrued considering the number of uncollected data for each data type and the current data processing capability of the device detected after recovery from the failure Was derived the collection method of the data, according to the derived the collection method, was provided and a third step of collecting the data of uncollected.

According to this data collection apparatus and method, after recovery from a failure, uncollected data can be collected by a collection method according to the priority of the current data type and the data processing capability of the current data collection apparatus. Therefore, the data required by the external device can be distributed as quickly as possible after the failure is recovered.

According to the present invention, a highly reliable data collection apparatus and method can be realized.

It is a block diagram which shows the whole structure of the data collection delivery system by 1st Embodiment. It is a conceptual diagram with which it uses for description of a data classification. FIG. 3 is a block diagram showing a flow of a series of processes when data including operation information is transmitted from a data collection source terminal to a data collection and distribution server in the data collection and distribution system of the first embodiment. FIG. 3 is a block diagram illustrating a flow of a series of processes when data including alert information is transmitted from a data collection source terminal to a data collection and distribution server in the data collection and distribution system according to the first embodiment. It is a block diagram which shows a mode when a failure generate | occur | produces in the data collection delivery server of 1st Embodiment. (A) is a block diagram for explaining the first problem of the data collection and distribution system after failure recovery, and (B) is a conceptual diagram for explaining the first problem. (A) is a block diagram for explaining the second problem of the data collection and distribution system after failure recovery, and (B) is a conceptual diagram for explaining the second problem. It is a conceptual diagram which shows the structural example of a terminal information table. It is a conceptual diagram which shows the structural example of an accumulation | storage table. It is a conceptual diagram which shows the structural example of the priority management table by 1st Embodiment. It is a conceptual diagram which shows the structural example of a terminal list group. It is a conceptual diagram which shows the structural example of a server state management table. It is a conceptual diagram which shows the structural example of a residual data number management table. It is a conceptual diagram which shows the structural example of the residual data collection system management table by 1st Embodiment. It is a conceptual diagram which shows the structural example of an operation information judgment reference file. It is a conceptual diagram which shows the structural example of an alert information determination reference | standard file. It is a flowchart which shows the process sequence of an operation information priority assignment process. It is a flowchart which shows the process sequence of a sub classification priority registration process. It is a flowchart which shows the process sequence of the residual data collection process by 1st Embodiment. It is a flowchart which shows the process sequence of a server state detection process. It is a flowchart which shows the process sequence of a 1st residual data number acquisition process. It is a flowchart which shows the process sequence of a 2nd remaining data number acquisition process. It is a flowchart which shows the process sequence of the residual data collection system derivation | leading-out process by 1st Embodiment. It is a flowchart which shows the process sequence of a residual data collection command production | generation process. It is a block diagram which shows the whole structure of the data collection delivery system by 2nd Embodiment. It is a block diagram which shows the flow of a series of processes when the data containing operation information are transmitted to the data collection delivery server from the data collection origin terminal in the data collection delivery system of 2nd Embodiment. It is a block diagram which shows the flow of a series of processes when the data containing alert information are transmitted to the data collection delivery server from the data collection origin terminal in the data collection delivery system of 2nd Embodiment. It is a block diagram which shows a mode when a failure generate | occur | produces in the data collection delivery server of 2nd Embodiment. It is a conceptual diagram which shows the structural example of the priority management table by 2nd Embodiment. It is a conceptual diagram which shows the structural example of the residual data collection system management table by 2nd Embodiment. It is a flowchart which shows the process sequence of the residual data collection process by 2nd Embodiment. It is a flowchart which shows the process sequence of the residual data number acquisition process by 2nd Embodiment. It is a flowchart which shows the process sequence of the residual data collection system derivation | leading-out process by 2nd Embodiment. It is a flowchart which shows the process sequence of a residual data acquisition process.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

(1) First Embodiment (1-1) Configuration of Data Collection / Distribution System According to this Embodiment In FIG. 1, reference numeral 1 denotes a data collection / distribution system according to this embodiment as a whole. The data collection / distribution system 1 includes a data collection source tenant 3 composed of a plurality of data collection source terminals 2 and one or a plurality of data collection / delivery servers 4 connected via a first network 5. The collection / delivery server 4 is connected to one or a plurality of external servers 7 via the second network 6.

Each data collection source terminal 2 is equipped with a sensor 10 (hereinafter referred to as one of a pressure sensor, a temperature sensor, and a vibration sensor), and the state of the measurement target (pressure, temperature) output from the sensor 10. Or a communication terminal device having a function of periodically transmitting measured values representing vibration) to the data collection / delivery server 4 as data in a predetermined format, and comprising a CPU (Central Processing Unit) 11 and a main storage device 12. The

The CPU 11 is a processor that controls the operation of the entire data collection source terminal 2. The main storage device 12 is composed of, for example, a semiconductor memory and is used as a work memory for the CPU 11. The main storage device 12 is provided with a buffer area 13 for temporarily storing measurement values output from the sensor 10. The main storage device 12 also stores a data communication control unit 14 that is a program having a function of periodically transmitting data stored in the buffer area 13 to the data collection / delivery server 4.

The data collection / delivery server 4 accumulates data transmitted from each data collection source terminal 2 of the data collection source tenant 3 and requests data from the accumulated data in response to a request from the external server 7. Is a server device having a function of delivering the message to the external server 7. The data collection / delivery server 4 includes a CPU 20, a main storage device 21, and an auxiliary storage device 22.

The CPU 20 is a processor that controls the operation of the entire data collection / delivery server 4. The main storage device 21 is composed of a semiconductor memory, for example, and is mainly used for holding various programs. Programs such as a data processing unit 30, a failure detection unit 31, an operation information priority detection unit 32, a server state detection unit 33, a residual data number acquisition unit 34, a residual data collection control unit 35, and a command delivery unit 36, which will be described later. It is stored and held in the main storage device 21. The main storage device 21 is also provided with a buffer area 37 for temporarily holding data transmitted from the data collection source terminal 2.

The auxiliary storage device 22 is used to hold various programs and information for a long period of time, and includes a large-capacity storage device such as a hard disk device or an SSD (Solid State Drive). The auxiliary storage device 22 includes a terminal information table 40, a storage table 41, a priority management table 42, a terminal list group 43, a server state management table 44, a residual data number management table 45, a residual data collection method management table 46, A specification information file 47, a search distribution log file 48, a high-speed distribution log file 49, an operation information determination reference file 50, and an alert information determination reference file 51 are also stored.

The external server 7 is a server device installed by a service user and having a function of requesting the data collection / delivery server 4 to distribute data necessary for business. The external server 7 includes a CPU 60, a main storage device 61, an input device 62, and an output device 63.

The CPU 60 is a processor that controls the operation of the entire external server 7. The main storage device 61 is composed of a semiconductor memory, for example, and is mainly used for holding various programs. An input / output control unit 64, which is a program for requesting the data collection / delivery server 4 to deliver necessary data, is also stored and held in the main storage device 61.

The input device 62 is hardware that is used by a service user to perform various operations, and includes, for example, a mouse and a keyboard. The output device 63 includes a monitor, a printer, and the like, and displays the processing result of the input / output control unit 64.

The first and

second networks

5 and 6 are constituted by, for example, the Internet, a LAN (Local Area Network), a SAN (Small Area Network), or the like. The first and

second networks

5 and 6 may be the same network.

FIG. 2 shows a correspondence relationship between each data collection source terminal 2 in the data collection source tenant 3 and the data type of data collected from these data collection source terminals 2. Data types of data collected by the data collection / delivery server 4 from each data collection source terminal 2 include alert information and operation information.

The alert information is information indicating that the measurement value of the sensor 10 (FIG. 1) included in the data transmitted from the data collection source terminal 2 indicates an abnormal value, and the operation information is the data collection source terminal 2. This is information representing the measured value of the pressure, temperature or vibration of the measurement object obtained based on the data transmitted from the data collection source terminal 2 during normal operation.

Also, the operation information can be classified into several types (hereinafter referred to as operation information sub-types or simply sub-types) according to the contents. In the present embodiment, as operation information sub-type, pressure information that is a measurement result of the sensor 10 when the sensor 10 mounted on the data collection source terminal 2 is a pressure sensor, and the sensor 10 is a temperature sensor. There are three types of temperature information, which is the measurement result of the sensor in the case, and vibration information, which is the measurement result of the sensor when the sensor 10 is a vibration sensor.

One data collection source terminal 2 is equipped with one type of sensor 10 among a pressure sensor, a temperature sensor, and a vibration sensor. Therefore, as shown in the column of [Terminal] in FIG. From the original terminal 2, in addition to data corresponding to the alert information, operation information corresponding to one type of operation information sub-type is transmitted to the data collection and distribution server 4 as data.

FIG. 3 shows a flow of a series of processing when the data transmitted from the data collection source terminal 2 in the data collection source tenant 3 to the data collection and distribution server 4 corresponds to the operation information. First, the output (measured value) of the sensor 10 of the data collection source terminal 2 is stored in the buffer area 13 of the main storage device 12 (FIG. 1) of the data collection source terminal 2 (SP1).

The measurement values stored in the buffer area 13 are periodically read by the data communication control unit 14 and transmitted to the data collection / delivery server 4 as data in a predetermined format (SP2). It is stored in the buffer area 37 of the storage device 21 (FIG. 1) (SP3).

Further, the data stored in the buffer area 37 of the data collection / delivery server 4 is periodically read out by the data processing unit 30 described later (SP4) and the operation information determination criterion described later stored in the operation information determination criterion file 50. And the operation information sub-type of the data based on the later-described alert information determination criteria stored in the alert information determination criteria file 51 and whether the data corresponds to the operation information or the alert information ( The type of the measured value stored in the data and whether or not the measured value is in the normal range as the measured value of the corresponding operation information sub type are determined (SP5). When it is determined by this determination that the data corresponds to the operation information (that is, when the measurement value stored in the data is determined to be within the normal range as the measurement value of the corresponding operation information sub-type) The data is registered in the accumulation table 41 (SP6).

Thereafter, when a data search / distribution request (hereinafter referred to as a data search / distribution request) for which conditions such as operation information sub-type and storage date / time are specified is given from the external server 7 to the data collection / delivery server 4 (SP7). The data processing unit 30 searches the storage table 41 for data that satisfies the conditions specified in the data search distribution request (SP8).

Then, the data detected by this search is distributed to the external server 7 that is the transmission source of the data search distribution request (SP9), the content of the data, the URL (Uniform Resource Locator) of the external server 7 that is the distribution destination, And history information such as delivery date and time is written in the search delivery log file 48 as a search delivery log (SP10).

FIG. 4 shows a flow of a series of processing when the data transmitted from the data collection source terminal 2 to the data collection / delivery server 4 corresponds to alert information. In this case, the processing from step SP20 to step SP24 is the same as the processing from step SP1 to step SP5 described above with reference to FIG.

If the data processing unit 30 determines in step SP24 that the data read from the buffer area 37 of the main storage device 21 (FIG. 1) corresponds to the alert information (that is, stored in the data). If it is determined that the measured value is in the abnormal range as the measured value of the corresponding operation information sub-type), the data is immediately distributed to the external server 7 (SP25), the contents of the data, and the URL of the external server of the distribution destination Then, history information such as delivery date and time is written in the high-speed delivery log file 49 as a high-speed delivery log (SP26).

On the other hand, FIG. 5 shows a state of the data collection source terminal 2, the data collection / distribution server 4, and the external server 7 when a failure has occurred in the data collection / delivery server 4 that cannot execute the data processing necessary for providing the service. FIG. 5 shows an example in which the failure that has occurred in the data collection / delivery server 4 is a failure in which no data from the data collection source terminal 2 or data retrieval / delivery request from the external server 7 can be received. .

In this case, since data cannot be transmitted to the data collection / delivery server 4 on the data collection source terminal 2 side, the data is stored in the buffer area 13 of the main storage device 12 (FIG. 1) by the sensor 10 and is transmitted by the data communication control unit 14. The sensor output read from the buffer area 13 is returned to the buffer area 13 (SP30, SP31). As a result, a large amount of sensor output (measured value information) continues to accumulate in the buffer area 13. In addition, since the external server 7 cannot transmit a data search / distribution request to the data collection / delivery server 4, it cannot acquire data necessary for business.

On the other hand, FIG. 6A shows a flow of a series of processes executed after the data collection / delivery server 4 is recovered from the above-described failure. In this case, on the data collection source terminal 2 side, the data communication control unit 14 sequentially reads a large amount of sensor output remaining in the buffer area 13 of the main storage device 12 (FIG. 1) (SP40), and collects data as data in a predetermined format. The data is sequentially transmitted to the distribution server 4 (SP41).

On the data collection / delivery server 4 side, the data processing unit 30 reads from the buffer area 37 the data transmitted from each data collection source terminal 2 and stored in the buffer area 37 of the main storage device 21 (FIG. 1). (SP42), step SP5 in FIG. 3 (step SP24 in FIG. 4) is determined (SP43), and the processing after step SP6 in FIG. 3 or the processing after step SP25 in FIG. 4 depending on the data type of the data. Execute.

That is, when the data read from the buffer area 37 corresponds to the operation information, the data processing unit 30 stores the data in the accumulation table 41 (SP44). Further, when a data search / distribution request is subsequently given from the external server 7 (SP45), the data processing unit 30 reads data satisfying the designated condition from the storage table 41 and distributes it to the external server 7 (SP46, SP47). The history information (search distribution log) is written in the search distribution log file 48 (FIG. 1). If the data corresponds to alert information, the data is immediately transmitted to the external server 7 (SP48), and the history information (high-speed distribution log) is stored in the high-speed distribution log file 49 (FIG. 1). Write.

(1-2) First Residual Data Collection Function According to this Embodiment By the way, in the series of processing after the failure recovery of the data collection / delivery server 4 described above with reference to FIG. The data (measured value) accumulated in the buffer area 13 of the main storage device 12 (FIG. 1) of each data collection source terminal 2 is, for example, a data type (hereinafter referred to as operation information) set at the start of service provision to a service user. Each data collection source terminal 2 in the data collection source tenant 3 is controlled so as to be collected with the priority of (including the sub type).

For this reason, for example, when the priority of the data type seen from the service user has changed since the start of service provision, the data of the data type with the current high priority seen from the service user is currently used by the service As a result, the data type that the current service user wants to acquire quickly cannot be acquired promptly after the failure of the data collection and distribution server. A situation may occur.

For example, as shown in FIG. 6B, when the priority set in the past is “4” for the data type “pressure”, the current priority viewed from the current service user is “2”. Even if it is, since data is collected according to priorities set in the past, data including operation information related to “pressure” is called “alert information”, “pressure”, “temperature” and “vibration” Among the data types, the data collection / delivery server 4 collects the last. As a result, even after the data collection / delivery server 4 is recovered from the failure, until the operation information of the data type “pressure” is transmitted from the corresponding data collection source terminal 2 to the data collection / delivery server 4, There is a possibility that the service user cannot acquire the “pressure” operation information and cause a business delay.

On the other hand, FIG. 7A, in which the same reference numerals are attached to the corresponding parts to FIG. 6A, is a degenerate performance that has deteriorated compared to normal operation after failure recovery, for example, as shown in FIG. 7B. The flow of processing of the data collection source terminal 2 and the data collection and distribution server 4 when the data collection and distribution server 4 is operated in the state is shown.

When data is transmitted from the data collection source terminal 2 to the data collection / distribution server 4 at the same frequency as during normal operation in a situation where the data collection / delivery server 4 is in a degenerated state, the data collection / distribution server 4 determines the transmission frequency. Since the data cannot be processed together, the data remaining on the buffer area 37 of the main storage device 21 (FIG. 1) of the data collection / delivery server 4 continues to increase monotonically as time elapses. The data from the data collection source terminal 2 may overflow (overflow) from the buffer area 27.

Normally, data overflowing from the buffer area 37 is returned to the data collection source terminal 2 of the transmission source (SP49), but how to handle the returned data is determined by the data communication control unit 14 of the data collection source terminal 2. Depends on the specifications. For this reason, depending on the specifications of the data communication control unit 14, the returned data may not be retransmitted to the data collection / delivery server 4, and as a result, the service user cannot acquire data necessary for the business. May also occur.

Therefore, in the data collection and distribution system 1 of the present embodiment, when the data collection and distribution server 4 recovers from the failure as described above with reference to FIG. 5, the uncollected data remaining in the data collection source tenant 3 at that time Data (hereinafter referred to as “residual data”), the order and timing of each appropriate data type in accordance with the priority of each current data type viewed from the service user and the current status of the data collection / delivery server 4 The data collection / delivery server 4 is equipped with a first residual data collection function for controlling each data collection source terminal 2 so that data can be collected at an appropriate frequency (the number of data per unit time).

In practice, in the case of the data collection / distribution system 1, the data collection / distribution server 4 operates based on the data search / distribution request given from the external server 7 during the most recent predetermined period (hereinafter referred to as one month). Based on the number of search delivery requests for each sub-type, the priority of each current operation information sub-type viewed from the service user is periodically detected and managed. At this time, the data collection / distribution server 4 assumes that the operation information sub-type having a larger number of search distribution requests in the most recent month has higher priority as viewed from the service user at the present time. Detect priority.

When the data collection / delivery server 4 recovers from the failure described above with reference to FIG. 5, the data collection / delivery server 4 first executes the performance measurement test using the performance measurement tool to Detect the data processing ability of

Further, the data collection / delivery server 4 inquires the individual data collection source terminal 2 about the number of data of each data type remaining in the data collection source terminal 2, and at that time, in the data collection source tenant 3 The number of data items for each data type remaining in is acquired.

The data collection / delivery server 4 determines the priority of each data type managed at that time, the current data processing capability, and the number of data items for each data type remaining in the data collection source tenant 3. Based on the data type, a residual data collection method (hereinafter referred to as a residual data collection method) for deriving the residual data in what order, timing, and frequency is derived. At this time, the data collection / delivery server 4 derives a residual data collection method for collecting alert information earlier (first) than the operation information.

Further, the data collection / delivery server 4 determines the data type of the residual data to be transmitted, the transmission timing of the residual data, and the transmission frequency when transmitting the residual data in accordance with the residual data collection method derived in this way. By generating a command for each designated data collection source terminal 2 (hereinafter referred to as a residual data collection command) and transmitting the generated residual data collection command to the corresponding data collection source terminal 2, Residual data is transmitted from each data collection source terminal 2 in the order, timing, and frequency according to the residual data collection method derived as described above.

As a means for realizing the first residual data collection function according to the present embodiment as described above, the main storage device 21 (FIG. 1) of the data collection and distribution server 4 includes the data processing unit 30, as described above. A failure detection unit 31, an operation information priority detection unit 32, a server state detection unit 33, a residual data number acquisition unit 34, a residual data collection control unit 35, and a command delivery unit 36 are stored.

The data processing unit 30 stores the data type of the data from the data collection source terminal 2 stored in the buffer area 37 (FIG. 1) of the main storage device 21 and will be described later stored in the operation information determination reference file 50 (FIG. 1). This is a program having a function of making a determination based on an operation information determination criterion and an alert information determination criterion described later stored in the alert information determination criterion file 51.

As described above, when the data processing unit 30 determines that the data type of the data is operation information in the determination processing, the data processing unit 30 registers the data in the accumulation table 41, while data search / delivery from the external server 7 In response to the request, data satisfying the condition is read from the accumulation table 41 and transmitted to the external server 7, and the history information (search distribution log) is stored in the search distribution log file 48 (FIG. 1). When the data processing unit 30 determines that the data is alert information in the determination process, the data processing unit 30 distributes the alert information to the external server 7 at a high speed and also stores the history information (high-speed distribution log) in a high-speed distribution log file. 49 (FIG. 1).

In addition, the failure detection unit 31 monitors the own data collection / delivery server 4 (hereinafter referred to as the own server), and notifies the system administrator of the occurrence of a failure in the own server or the time ( In the following, this is called a failure occurrence time).

Further, the operating information priority detection unit 32 has a function of periodically detecting the priority of each current data type as viewed from the service user based on the search distribution log stored in the search distribution log file 48. It is. The operating information priority detection unit 32 updates a priority management table 42 described later based on the detection result.

The server status detection unit 33 is a program having a function of detecting the current status (data processing capability) of the local server after the failure of the local server is recovered. The server state detection unit 33 registers and manages the detected current state of the own server in a server state management table 44 described later.

The residual data number acquisition unit 34 is a program having a function of acquiring the number of data for each data type remaining in the data collection source tenant 3 after the failure of the own server is recovered. The residual data number acquisition unit 34 stores and manages the number of data items for each acquired data type (alert information and each operation information subtype) in a residual data number management table 45 described later.

The residual data collection control unit 35 also detects the current priority for each data type detected by the operation information priority detection unit 32 and the current status of the own server detected by the server state detection unit 33 after the failure of the own server is recovered. Is a program having a function of deriving the above-described residual data collection method based on the state of the data and the number of residual data items for each data type in the data collection source tenant 3 acquired by the residual data number acquisition unit 34 . The residual data collection control unit 35 registers and manages the derived residual data collection method in a residual data collection method management table 46 described later. Further, the residual data collection control unit 35 sends the above-described residual data collection command for collecting data in the order, timing and frequency according to the residual data collection method registered in the residual data collection method management table 46 to the data collection source terminal. Generate every two.

The command delivery unit 36 is a program having a function of delivering a command to each data collection source terminal 2. The command delivery unit 36 delivers the residual data collection command for each data collection source terminal 2 generated by the residual data collection control unit 35 to the corresponding data collection source terminal 2.

On the other hand, in the auxiliary storage device 22 (FIG. 1) of the data collection / delivery server 4, as means for realizing the first residual data collection function of the present embodiment, a terminal information table 40, an accumulation table 41, a priority table are provided. Degree management table 42, terminal list group 43, server state management table 44, residual data number management table 45, residual data collection method management table 46, specification information file 47, search distribution log file 48, high-speed distribution log file 49, operation information A determination criterion file 50 and an alert information determination criterion file 51 are stored.

The terminal information table 40 is a table used for managing the data collection source terminal 2 in the data collection source tenant 3, and as shown in FIG. 8, a terminal ID column 40A, a model name column 40B, and a version column 40C. A buffer size column 40D and a communication terminal identification ID column 40E are provided. In the terminal information table 40, one row corresponds to one data collection source terminal 2.

The terminal ID column 40A stores an identifier (terminal ID) unique to each data collection source tenant 3 assigned to each data collection source terminal 2, and the model name column 40B stores the corresponding data collection source. The model name of the data collection software installed in the terminal 2 is stored.

The version column 40C stores the version of the corresponding data collection software, and the buffer size column 40D stores the buffer area 13 (defined in the main storage device 12 (FIG. 1) of the corresponding data collection source terminal 2 (FIG. 1). The maximum number of data items that can be stored is stored in FIG. Furthermore, an identifier (communication terminal identification ID) as a communication terminal assigned to the corresponding data collection source terminal 2 is stored in the communication terminal identification ID column 40E.

The accumulation table 41 is a table used for managing operation information obtained based on data transmitted from each data collection source terminal 2, and as shown in FIG. 9, the operation information ID column 41A, An operation information sub type column 41B, a terminal ID column 41C, a pressure column 41D, a temperature column 41E, a vibration column 41F, and an accumulation date / time column 41G are configured. In the accumulation table 41, one row corresponds to one piece of operation information.

In the operation information ID column 41A, an identifier (operation information ID) unique to the operation information assigned to each operation information by the data collection and distribution server 4 is stored, and the operation information sub type column 41B corresponds to the operation information sub type column 41B. Stores the name (subtype name) of the operation information. The terminal ID column 41C stores the terminal ID of the data collection source terminal 2 that is the transmission source of the corresponding operation information, and the accumulation date and time column 41G stores the date and time when the operation information is registered in the accumulation table 41. .

Furthermore, the pressure column 41D, the temperature column 41E, and the vibration column 41F store the values of the operation information included in the data transmitted from the corresponding data collection source terminal 2, respectively.

In the case of the present embodiment, in the data transmitted from the data collection source terminal 2 to the data collection / delivery server 4, there are fields corresponding to the respective operation information subtypes (hereinafter referred to as operation information fields). Is provided. Then, the data communication control unit 14 (FIG. 1) of the data collection terminal 2 stores the measurement value output from the sensor 10 in the operation information field associated with the corresponding operation information subtype in the operation information field. Then, data storing “NULL”, which means that no information exists in the other operation information fields, is generated, and the data is transmitted to the data collection / delivery server 4.

For example, the data communication control unit 14 of the data collection source terminal 2 in which a pressure sensor is mounted as the sensor 10 operates the measurement value output from the sensor 10 in association with “pressure” in the above-described operation information field. Data stored in the information field and “NULL” stored in the other operation information field is transmitted to the data collection / delivery server 4. In addition, the data communication control unit 14 of the data collection source terminal 2 in which the temperature sensor is mounted as the sensor 10 operates the operation information associated with the measured value output from the sensor 10 and “temperature” in the operation information field described above. The data stored in the field and stored with “NULL” in the other operation information field is transmitted to the data collection / delivery server 4. Further, the data communication control unit 14 of the data collection source terminal 2 on which the vibration sensor is mounted as the sensor 10 operates the operation value associated with the “vibration” in the operation information field described above. The data stored in the field and stored with “NULL” in the other operation information field is transmitted to the data collection / delivery server 4.

Therefore, the value stored in the operation information field corresponding to the “pressure” of the data transmitted from the data collection source terminal 2 is stored in the pressure column 41D of the accumulation table 41, and the data is stored in the temperature column 41E. The value stored in the operation information field corresponding to “Temperature” is stored, and the value stored in the operation information field corresponding to “Vibration” of the data is stored in the vibration column 41F.

The priority management table 42 is a table used for managing priorities for each operation information sub type, and as shown in FIG. 10, a priority column 42A, an operation information sub type column 42B, and a terminal list column 42C. And a search delivery execution count column 42D. In the priority management table 42, one row corresponds to one priority.

In the operation information sub type column 42B, the sub type name of each operation information sub type is stored, and in the priority column 42A, a numerical value indicating the priority of the corresponding sub type is stored. In the case of the present embodiment, as described above, the priority is higher as the numerical value stored in the priority column 42A is smaller. Therefore, in the example of FIG. 10, the priority of the operation information sub-type “DATA1 (pressure)” whose numerical value stored in the priority column 42A is “1” is the highest, and is stored in the priority column 42A. The operation information sub-type “DATA2 (temperature)” with the numerical value “2” has the next highest priority, and the operation information “DATA3 (vibration)” with the numerical value “3” stored in the priority column 42A. This indicates that the sub type has the lowest priority.

The terminal list column 42C stores the name (terminal list name) of the corresponding terminal list 43A in the terminal list 43A constituting the terminal list group 43 described later, and the search distribution execution count column 42D corresponds to the name. Stores the number of times the operation status sub-type data is distributed in response to a request from the external server 7.

As shown in FIG. 11, the terminal list group 43 includes a plurality of terminal lists 43A provided in correspondence with the respective operation information subtypes. In these terminal lists 43A, the terminal IDs of all the data collection source terminals 2 that transmit the data of the corresponding operation information sub type to the data collection and distribution server 4 (the sensor 10 for the corresponding operation information sub type is mounted). Are stored respectively.

The server status management table 44 is a table used for managing the current status of the data collection / delivery server 4 detected by the server status detection unit 33 (FIG. 1). As shown in FIG. A column 44A, a measurement time column 44B, a status column 44C, and a processing number column 44D are configured. In the server state management table 44, one row (entry) corresponds to one data collection / delivery server 4.

In the server name column 44A, a name (server name) unique to the data collection / delivery server 4 assigned to the data collection / delivery server 4 is stored, and in the measurement time column 44B, the corresponding data collection / delivery server 4 Stores the date and time when the data processing capability was measured. The status column 44C stores the ratio (%) of the data processing capacity of the corresponding data collection / delivery server 4 to the normal time at that date and time, and the processing number column 44D stores the current data collection / delivery server 4 current rate. Stores the number of data processing cases per unit time (hereinafter referred to as 1 minute).

Furthermore, the residual data number management table 45 is a table used for managing the number of data items for each data type remaining in the data collection source tenant 3 acquired by the residual data number acquisition unit 34 (FIG. 1). As shown in FIG. 13, a data type column 45A and a residual data number column 45B are provided.

In the data type column 45A, names of individual data types (“AREAT (alert information)”, “DATA1 (pressure)”, “DATA2 (temperature)”, “DATA3 (vibration)”) are stored, respectively, and residual data The number field 45B stores the number of data of the corresponding data type remaining in the data collection source tenant 3 acquired by the remaining data number acquisition unit 34.

The residual data collection method management table 46 is a table used for managing the above-described residual data collection method derived by the residual data collection control unit 35 (FIG. 1). As shown in FIG. A column 46A, a processing time column 46B, a transmission interruption time column 46C, a transmission interval column 46D, and a transmission data number column 46E are configured.

In the priority column 46A, the priority assigned to each data type is stored. The processing time column 46B stores the time (processing time) required to process all data of the data type to which the corresponding priority is assigned among the data remaining in the data collection source tenant 3.

Further, in the transmission interruption time column 46C, when data of each data type is transmitted from the data collection source tenant 3 to the data collection / delivery server 4 in the order according to the priority, data of the data type having the highest priority is transmitted. Stores the time (hereinafter referred to as “transmission interruption time”) from the start of transmission of data of the data type to which the corresponding priority is assigned.

Further, in the transmission interval column 46D, the data collection source terminal 2 that has received the above-mentioned residual data collection command sends a data interval of the data type specified in the residual data collection command to the data collection / delivery server 4 (this In the embodiment, “1 minute” is stored), and the number of data items of the corresponding data type to be transmitted at the time interval is stored in the transmission data number field 46E.

The specification information file 47 is a file in which information related to the specifications of the data collection / delivery server 4 is stored. Specifically, the specification information file 47 stores the number of data processing items per unit time (1 minute) when the data collection / delivery server 4 is in normal operation.

The search distribution log file 48 stores history information (search distribution log) when operating information satisfying the condition is transmitted to the external server 7 in response to a data search distribution request from the external server 7 as described above. The high-speed distribution log file 49 is a file for storing history information (high-speed distribution log) when alert information is transmitted to the external server 7.

In addition, the operation information determination standard file 50 is a determination criterion for each operation information sub type for determining which operation information sub type includes the operation information included in the data transmitted from the data collection source terminal 2 (hereinafter referred to as these Are referred to as operation information judgment criteria). Specifically, as shown in FIG. 15, in the operation information determination standard file 50, for each operation information sub type, the normal range of the measured value of the sensor 10 included in the data is the operation information determination type of the operation information sub type. Stored as

Further, the alert information determination criterion file 51 is a determination criterion (hereinafter referred to as an alert determination criterion) for determining whether or not the measurement value included in the data transmitted from the data collection source terminal 2 corresponds to the alert information. Called). Specifically, as shown in FIG. 16, the alert information determination reference file stores a range of measured values for each operation information subtype that should be determined to be abnormal.

(1-3) Various processes related to the first residual data collection function according to the present embodiment Next, various processes executed in the data collection / distribution server 4 in relation to the first residual data collection function according to the present embodiment Specific processing contents of the processing will be described. In the following description, the processing subject of various processes will be described as “program (...)”, But in practice, the process is executed by the CPU 20 (FIG. 1) based on the program. .

(1-3-1) Operation Information Priority Assignment Process FIG. 17 shows specific processing contents of the operation information priority assignment process periodically executed by the operation information priority detection unit 32 (FIG. 1). The operating information priority detection unit 32 follows the processing procedure shown in FIG. 17 to determine the priority of each operating information subtype registered in the priority management table 42 (FIG. 10) as viewed from the service user. Update to the priority of each operation information subtype.

Actually, when the operation information priority assignment process is started, the operation information priority detection unit 32 determines whether or not a record exists in the priority management table 42 (SP50). If the operation information priority detection unit 32 obtains a negative result in this determination, it proceeds to step SP52, and if it obtains a positive result, it deletes all the records existing in the priority management table 42 (SP51).

Subsequently, the operation information priority detection unit 32 acquires the search delivery log for the most recent month from the search delivery logs stored in the search delivery log file 48 (SP52). The operation information priority detection unit 32 sets the variable n to the initial value “1” (SP53), and thereafter, the operation information sub-corresponding to the nth operation information determination criterion from the operation information determination criterion file 50. The name (subtype name) of the type (hereinafter referred to as the target operation information subtype) is acquired (SP54).

Next, the operation information priority detection unit 32 extracts a search distribution log related to the target operation information subtype from the search distribution logs for the latest one month acquired in step SP52 (SP55), and extracts the search distribution log into the extracted search distribution log. Based on the target operation information sub-type data, the number of times per day that the data search distribution request is received from the external server 7 and the average value thereof are calculated (SP56).

Further, the operation information priority detection unit 32 acquires the terminal list name of the terminal list 43A (FIG. 11) associated with the target operation information subtype of the subtype name acquired in step SP54 (SP57). Further, the operation information priority detection unit 32 assigns a priority to the operation information sub type for which the sub type name has been acquired in step SP54 and registers the operation information sub type in the priority management table 42. Execute (SP58), and then increase the variable n by “1” (SP59).

Furthermore, the operation information priority detection unit 32 determines whether or not the nth operation information determination criterion exists in the operation information determination criterion file 50 (SP60). If the operation information priority detection unit 32 obtains an affirmative result in this determination, the operation information priority detection unit 32 returns to step SP54, and thereafter performs the same processing for all the operation information determination criteria registered in the operation information determination criteria file 50. The processing from step SP54 to step SP60 is repeated until the processing is completed.

Then, the operation information priority detection unit 32 eventually obtains an affirmative result in step SP60 by completing the processing of steps SP55 to SP58 for all the operation information determination criteria registered in the operation information determination criteria file 50. This operation information priority assignment process is terminated.

Note that the specific processing contents of the sub-type priority registration processing executed in step SP58 of the above-described operation information priority assignment processing are shown in FIG.

When the operation information priority detection unit 32 proceeds to step SP58 of the operation information priority assignment process, it starts the sub type priority registration process shown in FIG. The target operation information sub-type for which the type name has been acquired is provisionally registered in the priority management table 42 (FIG. 10) (SP70).

Specifically, the operation information priority detection unit 32 secures one unused row in the priority management table 42, sets “0” in the priority column 42A of the row, and the operation information sub type column 42B of the row. The sub-type name acquired at step SP54 of the operation information priority assignment process, the terminal list column 42C of the row, the terminal list name acquired at step SP57 of the operation information priority assignment processing, and the search delivery execution count column 42D of the row Each of the average values calculated in step SP56 of the operation information priority assignment process is stored.

Subsequently, the operation information priority detection unit 32 sets the variable t to the initial value “1” (SP71), and thereafter, among the records of the priority management table 42, the priority stored in the priority column 42A. It is determined whether or not there is a record having the same value as the variable t at that time (SP72).

If the operation information priority detection unit 32 obtains a negative result in this determination, the value of the variable t at that time is set as the priority of the target operation information sub-type provisionally registered in the priority management table 42 in step SP70. Assign (SP78). That is, the operation information priority detection unit 32 updates the numerical value (here, “0”) stored in the priority column 42A of the target operation information subtype in the priority management table 42 to the value of the variable t at that time. To do. Then, the operation information priority detection unit 32 thereafter ends this sub-type priority registration process.

On the other hand, when the operation information priority detection unit 32 obtains a positive result in the determination at step SP72, the operation information sub-type data to which the same priority as the variable t at that time is assigned per minute to get the average retrieval and distribution number _{N t} of (SP73). Specifically, the operation information priority detection unit 32 stores the search distribution execution count column 42D of the record in which the same numerical value as the value of the variable t is stored in the priority column 42A of the priority management table 42. _Is obtained as the number of times of search distribution Nt.

In addition, the operation information priority detection unit 32 acquires the average number of times of search distribution N ₀ per minute for the data of the target operation information sub-type provisionally registered in the priority management table 42 in step SP70 (SP74). . Specifically, the operation information priority detection unit 32 uses the value stored in the search distribution execution count field 42D of the record corresponding to the target operation information subtype provisionally registered in step SP70 in the priority management table 42, as follows. to get as consuming retrieval and distribution number N _0.

Thereafter, the operation information priority detector 32 searches delivery count _{N 0} obtained in step SP74 is equal to or greater than obtained search delivery count _{N t} in step SP73 (SP75).

Obtaining a negative result in this determination is that the number of search distributions of the operation information subtype whose priority value is set to the same value as the variable t at that time in the priority management table 42 is the target operation information subtype. Therefore, the priority of the target operation information sub-type is higher than the priority of the operation information sub-type whose priority value is set to the same value as the variable t at that time (priority level). This means that you should set a smaller value.

Thus, at this time, the operation information priority detection unit 32 increases the value of the variable t by “1” (SP76), returns to step SP72, and thereafter executes the processing after step SP72.

On the other hand, obtaining an affirmative result in the determination at step SP75 is that, in the priority management table 42, the search and distribution of the operation information subtype whose priority value is set to the same value as the variable t at that time. The number of times is less than the number of search distributions of the target operation information sub-type. Therefore, the priority of the operation information sub-type in which the priority of the target operation information sub-type is set to the same value as the variable t at that time. This means that it should be set lower than the degree (larger value indicating the priority).

Thus, at this time, the operation information priority detection unit 32 currently prioritizes all operation information subtypes whose priority value is set to a value larger than the value of the variable t at that time in the priority management table 42. The priority management table 42 is updated so that each degree value is incremented by “1” (SP77).

Also, the operation information priority detection unit 32 assigns the value of the variable t at that time as the priority of the target operation information sub-type provisionally registered in the priority management table 42 in step SP70 (SP78). That is, the operation information priority detection unit 32 updates the numerical value (here, “0”) stored in the priority column 42A of the target operation information subtype in the priority management table 42 to the value of the variable t at that time. To do. Then, the operation information priority detection unit 32 thereafter ends this sub-type priority registration process.

(1-3-2) Residual Data Collection Processing (1-3-2-1) Flow of Residual Data Collection Processing On the other hand, in FIG. A flow of residual data collection processing executed in the data collection / delivery server 4 after recovery from the failure is shown.

In the data collection / delivery server 4, when recovering from such a failure, first, the server state detection unit 33 (FIG. 1) executes server state detection processing for detecting the current server state of the data collection / delivery server 4 (SP80). . The server state detection unit 33 detects the “current server state” as a ratio of the processing capacity of the current data collection / delivery server 4 to the processing capacity of the data collection / delivery server 4 during normal operation as described above.

Subsequently, the residual data number acquisition unit 34 (FIG. 1) performs a residual data number acquisition process for acquiring the number of uncollected data (residual data) currently remaining in the data collection source tenant 3 for each data type. Execute (SP81).

Next, the residual data collection control unit 35 (FIG. 1) calculates the priority of each current operation information subtype viewed from the service user registered in the priority management table 42 at that time, and is calculated in step SP80. Based on the current server status of the own server and the number of data items of residual data for each data type in the data collection source tenant 3 acquired in step SP81, the order, timing, and frequency for each data type for collecting residual data The residual data collection method deriving process for deriving the residual data collection method that defines the above is executed (SP82).

Further, the residual data collection control unit 35 provides the above-mentioned residual data collection for instructing each data collection source terminal 2 to transmit the residual data in the order, timing and frequency according to the residual data collection method derived in step SP82. Residual data collection command generation processing for generating a command for each data collection source terminal 2 is executed (SP83).

Thereafter, the command delivery unit 36 (FIG. 1) transmits each residual data collection command generated by the residual data collection control unit 35 as described above to the corresponding data collection source terminal 2, thereby collecting the data. All the remaining data remaining in the original tenant 3 is sequentially transmitted to the own server (SP84).

As described above, all the data accumulated in each data collection source terminal 2 during the failure of the data collection / delivery server 4 is collected by the data collection / delivery server 4, and necessary information of these data is stored in the accumulation table 41. Thereafter, the residual data collection process ends.

(1-3-2-2) Server Status Detection Processing FIG. 20 shows a specific processing procedure of the server status detection processing executed by the server status detection unit 33 in step SP80 of the residual data collection processing described above with reference to FIG. Show.

When the own server recovers from the failure, the server status detection unit 33 starts the server status detection process shown in FIG. 20, and first determines whether or not a record exists in the server status management table 44 (FIG. 12). (SP90). If the server state detection unit 33 obtains a negative result in this determination, it proceeds to step SP92, and if it obtains a positive result, it deletes all the records existing in the server state management table 44 (SP91).

Subsequently, the server state detection unit 33 obtains the data processing capability (per minute during normal operation) of the own server from the specification information file 47 (FIG. 1) stored in the auxiliary storage device 22 (FIG. 1). (Data processing number) is acquired (SP92).

Next, the server state detection unit 33 performs a performance measurement test for measuring the current data processing capability of the server itself (SP93). Specifically, the server state detection unit 33 activates a performance measurement software (not shown) given in advance and continues to provide data from the pseudo data collection source terminal 2 to the data processing unit 30 for one minute. Is processed by the data processing unit 30.

Thereafter, the server state detection unit 33 acquires a performance measurement result log file (not shown) created by the performance measurement software as a processing result of the data processing unit 30 during the one-minute test period (SP94). Also, the server state detection unit 33 acquires the total number of data processed during the test period stored in the performance measurement result log file as the current processing capacity of the own server (SP95).

Subsequently, the server status detection unit 33 determines the current server's current data processing capability based on the data processing capability of the server itself during normal operation acquired in step SP92 and the data processing capability of the current server acquired in step SP95. The state is calculated (SP96). Specifically, the server state detection unit 33 divides the current data processing capability of the local server acquired in step SP95 by the data processing capability of the local server during normal operation acquired in step SP92, and 100 is added to the result of the division. Is calculated as a ratio (%) of the current data processing capacity of the local server to the data processing capacity of the local server during normal operation.

Subsequently, the server status detection unit 33 acquires the server name of the server stored in advance (SP97), and registers the current server status in the server status management table 44 (SP98). Specifically, the server state detection unit 33 secures one unprocessed row in the server state management table 44, and stores the server name of the server in the server name column 44A of that row and the current time in the measurement time column 44B of that row. The current state of the current server calculated in step SP96 is stored in the time field 44C of the row, and the current number of data processing cases of the own server acquired in step SP95 is stored in the processing number column 44D of the row.

Then, the server state detection unit 33 thereafter ends this server state detection process.

(1-3-2-2-3) Residual Data Number Acquisition Process (1-3-2-2-3-1) First Residual Data Number Acquisition Process FIG. 21 is a flowchart of the residual data collection process described above with reference to FIG. An example of residual data number acquisition processing (hereinafter referred to as first residual data number acquisition processing) executed by the residual data number acquisition unit 34 (FIG. 1) is shown.

The first residual data number acquisition processing method shown in FIG. 21 causes each data collection source terminal 2 to calculate the current number of residual data for each data type, and notifies the data collection and distribution server 4 of the calculation result. In the data collection / delivery server 4, the total number of residual data in the data collection source tenant 3 is determined for each data type based on the number of residual data items for each data type notified from each data collection source terminal 2. It is a method to acquire each.

In practice, when the first residual data number acquisition processing method is applied as the residual data collection processing method, the residual data number acquisition unit 34 first stores a record in the residual data number management table 45 (FIG. 13). It is determined whether or not it exists (SP100). If the residual data number acquisition unit 34 obtains a negative result in this determination, it proceeds to step SP102, and if it obtains a positive result, it deletes all the records existing on the residual data number management table 45 (SP101).

Subsequently, the residual data number acquisition unit 34 sets the variable n to the initial value “1” (SP102), and thereafter, the terminal ID column 40A (in the record of the nth row of the terminal information table 40 (FIG. 8)) ( The terminal ID stored in FIG. 8) is acquired (SP103).

Next, the residual data count acquisition unit 34 includes all the operation information determination criteria stored in the operation information determination criteria file 50 (FIG. 15) and the alert information determination criteria stored in the alert information determination criteria file 51 (FIG. 16). Are acquired (SP104).

The residual data count acquisition unit 34 generates alert information based on the number of data determined to correspond to the operation information based on each operation information determination criterion acquired in step SP104 and the alert information determination criterion acquired in step SP104. A command for notifying the number of data determined to be equivalent (hereinafter referred to as an alert / operation information acquisition command) is generated (SP105).

Then, the remaining data number acquisition unit 34 transmits the alert / operation information acquisition command created in step SP105 to the command distribution unit 36 (FIG. 1) so as to transmit it to the data collection source terminal 2 assigned with the terminal ID acquired in step SP103. An instruction is given (SP106). As a result, the alert / operation information acquisition command is transmitted to the data collection source terminal 2 by the command delivery unit 36.

The data collection source terminal 2 that has received this alert / operation information acquisition command, based on the operation information determination criteria for each operation information subtype included in the alert / operation information acquisition command, and the alert information determination criteria, The data type of each data remaining in the buffer area 13 (FIG. 1) of its own main storage device 12 (FIG. 1) is determined, and corresponds to the number of data for each operation information sub-type and alert information. The number of data items is acquired, and the acquisition result is transmitted to the data collection and distribution server 4 together with the terminal ID of the own data collection source terminal 2.

Thereafter, the residual data number acquisition unit 34 increases the value of the variable n by “1” (SP107), and determines whether or not the record in the n-th row exists in the terminal information table 40 (FIG. 8) ( SP108). The residual data number acquiring unit returns to step SP when obtaining a positive result in this determination, and thereafter repeats the processing of step SP to step SP until obtaining a positive result at step SP.

Then, the residual data number acquisition unit 34 eventually transmits the above alert / operation information acquisition command to all the data collection source terminals 2 whose terminal IDs are registered in the terminal information table 40, and thereby returns a negative result in step SP108. Is obtained, a variable [sub-type name] _C for each operation information sub-type and a variable ARERT_C for alert information are respectively generated, and these values are set to initial values “0” (SP109). Note that “variable [sub-type name] _C” is a variable stored in the place where the sub-type name of the corresponding operation information sub-type is “[sub-type name]”.

Subsequently, the residual data number acquisition unit 34 sets the variable n to an initial value “0” (SP110), and sets the terminal ID stored in the terminal ID column 40A of the record in the n-th row of the terminal information table 40. Obtain (SP111). In addition, the remaining data number acquisition unit 34 receives the alert / data received from the data collection source terminal 2 in response to the alert / operation information acquisition command transmitted to the data collection source terminal 2 to which the terminal ID acquired in step SP111 is assigned. The execution result of the operation information acquisition command (the number of data items for each operation information sub type and the number of data items of alert information) is acquired (SP112).

The residual data count acquisition unit 34 adds the data count for each operation information subtype acquired in step SP112 and the data count of alert information to the corresponding variable [subtype name] _C or variable ARERT_C. The values of the values of each variable [sub-type name] _C and variable ARERT_C are updated (SP113).

Next, the residual data number acquisition unit 34 increases the variable n by “1” (SP114), and then determines whether or not the record of the nth row exists in the terminal information table 40 (SP115). The residual data number acquiring unit 34 returns to step SP111 when obtaining a positive result in this determination, and thereafter repeats the processing of step SP111 to step SP115 until obtaining a positive result at step SP115.

When the residual data count acquisition unit 34 eventually completes the processing of steps SP111 to SP113 for all the data collection source terminals 2 whose terminal IDs are registered in the terminal information table 40, it obtains a positive result in step SP115. The values of each variable [sub-type name] _C and variable ARERT_C are registered in the residual data number management table 45 (FIG. 13) (SP116).

Specifically, the residual data number acquisition unit 34 sets the value of each variable [sub-type name] _C to the corresponding operation information sub-type in the residual data number management table 45 (data type column 45A (FIG. 13)). Is stored in the remaining data number column 45B (FIG. 13), and the value of the variable ARERT_C is stored in the row corresponding to the alert information in the remaining data number management table 45 (data type column 45A). In the remaining data number column 45B).

Then, when the above process is completed, the residual data number acquisition unit 34 ends the first residual data number acquisition process.

(1-3-2-2-3-2) Second Residual Data Number Acquisition Process On the other hand, FIG. 22 shows the residual data executed by the residual data number acquisition unit 34 in step SP81 of the residual data collection process described above with reference to FIG. Another example of the number acquisition process (hereinafter referred to as a second residual data number acquisition process) is shown.

In the second residual data number acquisition processing method shown in FIG. 22, the data collection / delivery server 4 side calculates the number of data processing cases per unit time (1 minute) during normal operation for each data type. Based on the calculation result, the number of residual data items for each data type existing in the current data collection source tenant 3 is obtained by estimation.

In practice, when the second residual data number acquisition processing method is applied as the residual data collection processing method, the residual data number acquisition unit 34 first stores a record in the residual data number management table 45 (FIG. 13). It is determined whether or not it exists (SP120). If the residual data number acquisition unit 34 obtains a negative result in this determination, it proceeds to step SP122, and if it obtains a positive result, it deletes all records existing on the residual data number management table 45 (SP121).

Subsequently, the residual data number acquisition unit 34 acquires the high-speed distribution log for the latest several minutes from the search distribution log stored in the high-speed distribution log file 49 (FIG. 1) (SP122). In the case of the present embodiment, the remaining data number acquisition unit 34 acquires a high-speed delivery log whose output date and time is from 12:00 to 12:05 on the previous day in this step SP122.

Next, the residual data number acquisition unit 34 calculates the average value of the number of alert information processed per minute based on the acquired high-speed delivery log (SP123). Specifically, the residual data number acquisition unit 34 divides the high-speed delivery log acquired in step SP122 into five groups per minute (the output date and time is 12: 0, 12: 1, 12: 2, 5 groups of 12: 3 and 12: 4). The residual data count acquisition unit 34 counts the number of high-speed delivery logs for each group, and calculates the average count value for each group (that is, the average value of the number of processing data of alert information for 1 minute). To do.

Thereafter, the remaining data number acquisition unit 34 distributes to the external server 7 from the accumulation table 41 (FIG. 9) within the same period as the above-described predetermined period (5 minutes from 12:00 pm to 12:05 pm the previous day). The record information of all the records related to the operation information (records whose accumulation date and time is between 12:00 p.m. and 12: 5 p.m. the previous day) is acquired (SP124).

Then, the residual data number acquisition unit 34 calculates the average value of the number of data processing cases per operation information subtype per minute based on the record information acquired in step SP124 (SP125). Specifically, the residual data number acquisition unit 34 stores the record information acquired in step SP124 for each of the operation information subtypes in five groups per minute (accumulation date is 12: 0, 12: 1, 5 groups of 12: 2 range, 12: 3 range, and 12: 4 range). The remaining data count acquisition unit 34 counts the number of data of the operation information of each group for each minute for each operation information subtype, and averages the count values of each group for each minute (that is, the operation information) The average value of the number of processed data items per minute for each information sub type is calculated (SP125).

Subsequently, the remaining data number acquisition unit 34 acquires the failure occurrence time of the failure that occurred in the server from the failure detection unit 31 (FIG. 1) (SP126), and acquires the current time from a timer (not shown) (SP127). By subtracting the failure occurrence time acquired in step SP126 from the current time acquired in step SP127, the elapsed time from the start of data remaining in the data collection source tenant 3 due to the failure of the own server (from the failure occurrence time to the current (Hereinafter, referred to as data remaining time) is calculated (SP128).

Next, the remaining data number acquisition unit 34 performs multiplication of the average value of the number of processed data items of alert information per minute calculated in step SP123 and the data remaining time calculated in step SP128, thereby obtaining a data collection source. An estimated value of the number of remaining data in the alert information in the tenant 3 is calculated (SP129), and the calculated estimated value of the number of remaining data in the alert information is registered in the remaining data number management table 45 (FIG. 13) (SP130). Specifically, the residual data number acquisition unit 34 stores the estimated value of the residual data number of the alert information calculated in step SP129 in the row corresponding to the alert information in the residual data number management table 45 (data type column 45A (FIG. 13)). The data is stored in the remaining data number column 45B (FIG. 13) of “ARERT”.

Further, the residual data number acquisition unit 34 multiplies the average value of the number of processing data cases for each operation information subtype acquired in step SP125 by the data residual time calculated in step SP128, respectively. An estimated value of the number of remaining data items for each operation information subtype in the data collection source tenant 3 is calculated (SP131).

The residual data number acquisition unit 34 registers the estimated value of the residual data number for each operation information subtype calculated in step SP131 in the residual data number management table 45 (SP132). Specifically, the residual data number acquisition unit 34 sets the estimated value of the residual data number for each operating information subtype calculated in step SP131 to a row corresponding to the corresponding operating information subtype in the residual data number management table 45 ( The corresponding sub type name is stored in the remaining data number column 45B (FIG. 13) of the data type column 45A (line stored in FIG. 13).

Then, the residual data number acquisition unit 34 thereafter ends the second residual data number acquisition process.

(1-3-2-2-4) Residual data collection method derivation process FIG. 23 shows a residual data collection method executed by the residual data collection control unit 35 (FIG. 1) in step SP82 of the residual data collection process described above with reference to FIG. A specific processing procedure of the derivation processing is shown.

When the residual data collection control unit 35 starts the residual data collection method derivation process shown in FIG. 23, it first determines whether or not a record exists in the residual data collection method management table 46 (SP140). If the residual data collection control unit 35 obtains a negative result in this determination, it proceeds to step SP142, and if it obtains a positive result, it deletes all the records existing in the residual data collection method management table (SP141).

Subsequently, the residual data collection control unit 35 acquires the current data processing capability of the own server from the server state management table 44 (FIG. 12) (SP142). Specifically, the residual data collection control unit 35 stores the value (current number of data processing per minute) stored in the processing number column 44D (FIG. 12) of the row corresponding to the own server in the server state management table 44. To get.

Next, the residual data collection control unit 35 acquires the number of residual data of the alert information remaining in the data collection source tenant 3 at that time from the residual data number management table 45 (FIG. 13) (SP143). Specifically, the residual data collection control unit 35 includes the residual data number column 45B in the row corresponding to the alert information in the residual data number management table 45 (the row in which “ARERT” is stored in the data type column 45A (FIG. 13)). The value stored in (FIG. 13) is acquired.

Furthermore, the residual data collection control unit 35 divides the number of residual data acquired in step SP143 by the current data processing capability of the local server acquired in step SP142, thereby obtaining the data collection source tenant with the current data processing capability of the local server. 3 calculates the time (processing time) required to process all the alert information remaining in SP3 (SP144).

The residual data collection control unit 35 then registers the alert information processing time and transmission interruption time in the residual data collection method management table 46 (FIG. 14) (SP145). Specifically, the residual data collection control unit 35 secures an unused line on the residual data collection method management table 46, sets “0” in the priority column 46A (FIG. 14) of that line, and the processing time of that line. The processing time calculated in step SP144 is stored in the column 46B (FIG. 14), and “0” is stored in the transmission interruption time column 46C (FIG. 14) of that row.

Subsequently, the residual data collection controller 35 sets the variable m to the initial value “1” (SP146). Further, the residual data collection control unit 35 acquires the number of residual data items of the operation information sub type with the priority “m” remaining in the data collection source tenant 3 (SP147). Specifically, the residual data collection control unit 35 refers to the priority management table 42 (FIG. 10), and the operation information sub-record of the record whose value stored in the priority column 42A (FIG. 10) is “m”. The sub type name stored in the type column 42B is acquired, and among the records of the residual data number management table 45 (FIG. 13), the above-mentioned sub type name is stored in the data type name column 45A (FIG. 13). The value stored in the data number column 45B (FIG. 13) is acquired.

Next, the residual data collection control unit 35 divides the number of residual data acquired in step SP147 by the current data processing capability of the local server acquired in step SP142, so that the priority remaining in the data collection source tenant 3 is obtained. The time (processing time) required to process all the data of the operation information sub-type having the degree “m” with the current data processing capability of the local server is calculated (SP148).

Further, the residual data collection control unit 35 calculates the transmission interruption time of the data of the operation information sub type with the priority “m” (SP149). Specifically, the residual data collection control unit 35 selects all of the records registered in the residual data collection method management table 46 (FIG. 14) whose priority stored in the priority column 46A is less than “m”. The total value of the values stored in the record processing time column 46B is calculated as the transmission interruption time of the data of the operation information sub type with the priority “m”.

Thereafter, the residual data collection control unit 35 secures an unused line on the residual data collection method management table 46, “m” in the priority column 46A of that line, and the step in the processing time column 46B of that line. The processing time acquired in SP148 and the transmission interruption time acquired in step SP149 are stored in the transmission interruption time column 149 of the row (SP150).

Thereafter, the residual data collection control unit 35 increases the value of the variable m by “1” (SP151), and then refers to the priority management table 42 (FIG. 10) to determine the value of the variable m at that time. It is determined whether there is an operation information sub type assigned as a priority (SP152). If the residual data collection control unit 35 obtains a negative result in this determination, it returns to step SP147, and thereafter repeats the processing of step SP147 to step SP152 until a positive result is obtained in step SP152.

Through the processing of step SP142 to step SP152, the data collection source terminal 2 for each data type (alert information and each operation information subtype) taking into account the current data processing capability of the server itself is sent to the data collection / delivery server 4. The data transmission timing is derived respectively.

When the residual data collection control unit 35 eventually obtains a positive result in step SP by deriving the transmission timing of data from the data collection source terminal 2 to the data collection / delivery server 4 for each data type, the terminal information table 40 is obtained. (FIG. 8) The total number of data collection source terminals 2 in the data collection source tenant 3 (that is, the total number of data collection source terminals from which alert information is collected) is obtained by counting the total number of records in the above (SP153). .

Further, the residual data collection control unit 35 considers the current state of its own server, and the number of alert information data to be transmitted from each data collection source terminal 2 in one minute (hereinafter referred to as alert information transmission frequency). Is calculated (SP154). Specifically, the residual data collection control unit 35 divides the number of residual data items of the alert information acquired in step SP143 by the total number of terminals calculated in step SP153, and further divides the result of the residual data collection method management table 46 ( By dividing by the value stored in the processing time column 46B of the record corresponding to the alert information in FIG. 14), the transmission frequency of the alert information is calculated.

The residual data collection control unit 35 stores the alert information transmission frequency calculated in this way in the transmission data number field 46E (FIG. 14) of the corresponding record in the residual data collection method management table 46, and at the same time, the transmission interval field. “1” is stored in 46D (FIG. 14) (SP155).

Subsequently, the residual data collection controller 35 sets the variable x to the initial value “1” (SP156).

Further, the residual data collection control unit 35 acquires the terminal list 43A (FIG. 11) corresponding to the operation information sub-type whose priority is set to “x” in the priority management table 42 (FIG. 10) at that time ( SP157). Specifically, the residual data collection control unit 35 determines from the priority management table 42 the terminal list of the terminal list 43A stored in the terminal list field 42C of the record whose value stored in the priority field 42A is “x”. The terminal list 43A to which the terminal list name is assigned is acquired from the terminal list group 43 (FIG. 11).

Next, the residual data collection control unit 35 transmits the total number of data collection source terminals 2 registered in the terminal list 43A acquired in step SP157 (that is, data of operation information subtype data with a priority “x”). The total number of original terminals 2) is acquired (SP158). Further, the residual data collection control unit 35 acquires the number of data of the operation information sub type with the priority “x” remaining in the data collection source tenant 3 from the residual data number management table 45 (FIG. 13) ( SP159).

Then, the residual data collection control unit 35 considers the current state of its own server, and the number of data of the operation information sub-type with the priority “x” that each data collection source terminal 2 should transmit in one minute (hereinafter referred to as this) Is called the operating information transmission frequency of the operating information sub-type with the priority “x”) (SP160). Specifically, the residual data collection control unit 35 divides the residual data number of the data of the operation information sub type with the priority “x” acquired in step SP159 by the total number of terminals acquired in step SP158, and the division result Is further divided by the value stored in the processing time column 46B (FIG. 14) of the record corresponding to the operation information subtype with the priority “x” in the residual data collection method management table 46 (FIG. 14), The operation information transmission frequency of the operation information subtype with the priority “x” is calculated.

Then, the residual data collection control unit 35 determines the transmission frequency of the operation information of the operation information subtype with the priority “x” calculated in this way in the corresponding record of the residual data collection method management table 46 (FIG. 14). The data is stored in the transmission data number column 46E (FIG. 14), and “1” is stored in the transmission interval column 46D (FIG. 14) (SP161).

Thereafter, the residual data collection control unit 35 increases the value of the variable x by “1” (SP162), and then refers to the priority management table 42 (FIG. 10) to determine the value of the variable x at that time. It is determined whether or not there is an operation information sub type assigned as a priority (SP163). If the residual data collection control unit 35 obtains a negative result in this determination, it returns to step SP157, and thereafter repeats the processing of step SP157 to step SP163 until a positive result is obtained in step SP163.

Through the processing of step SP153 to step SP163 described above, the data collection source terminal 2 for each data type (alert information and each operation information sub-type) is considered from the data collection / delivery server 4 in consideration of the current state of the data collection / delivery server 4. Are respectively derived. Note that this transmission frequency can be said to be the frequency of residual data for each data type that can be collected by the data collection and distribution server 4 in consideration of the current state of the data collection and distribution server 4.

When the residual data collection control unit 35 eventually obtains a positive result in step SP163 by completing the processing of steps SP157 to SP161 for all priorities assigned to the operation information subtype, the residual data collection method is derived. End the process.

(1-3-2-5) Residual Data Collection Command Generation Processing FIG. 24 shows the specifics of the residual data collection command generation processing executed by the residual data collection control unit 35 in step SP83 of the residual data collection processing described above with reference to FIG. A typical processing procedure is shown.

When the residual data collection control unit 35 ends the residual data collection method derivation process described above with reference to FIG. 23, the residual data collection command generation process shown in FIG. 24 is started. First, the alert information determination reference file 51 (FIG. 16). The alert information judgment criteria stored in the are acquired (SP170).

Subsequently, the residual data collection control unit 35 sets the variable n to the initial value “1” (SP171), and thereafter, the terminal ID field of the record in the “n” line of the terminal information table 40 (FIG. 8). The terminal ID stored in 40A (FIG. 8) is acquired (SP172).

Next, the residual data collection control unit 35 generates the above-described residual data collection command with the data collection source terminal 2 to which the terminal ID acquired in step SP172 is assigned as the transmission destination (SP173). At this stage, information on the corresponding data collection method derived by the residual data collection method derivation process described above with reference to FIG. 23 is not registered in this residual data collection command.

Further, the residual data collection control unit 35 acquires the alert information collection method derived by the residual data collection method derivation process described above with reference to FIG. 23 (SP174). Specifically, the residual data collection control unit 35 determines the transmission interruption time of the record whose priority stored in the priority column 46A (FIG. 14) is “0” from the residual data collection method management table 46 (FIG. 14). The transmission interruption time stored in the column 46C (FIG. 14), the transmission interval stored in the transmission interval column 46D (FIG. 14) of the record, and the transmission data number column 46E (FIG. 14) of the record The number of data is acquired as the alert information data collection method.

Then, the residual data collection control unit 35 registers the alert information data collection method for the residual data collection command generated in step SP173 (SP175). Specifically, the residual data collection control unit 35 sets alert information as a data type to be targeted at each corresponding position in the residual data collection command, and uses the transmission interruption time acquired in step SP174 as the transmission interruption time. As the transmission frequency, a value obtained by dividing the number of data acquired in step SP174 by the transmission interval acquired in step SP174 is set.

Subsequently, the residual data collection controller 35 sets the variable “t” to the initial value “1” (SP176), and stores it in the priority column 42A (FIG. 10) from the priority management table 42 (FIG. 10). The sub type name stored in the operation information sub type column 42B (FIG. 10) of the record having the priority “t” is acquired (SP177).

Next, the residual data collection control unit 35 acquires the terminal list 43A (FIG. 11) corresponding to the operation information subtype to which the subtype name acquired in step SP177 is assigned (SP178), and puts it on the acquired terminal list 43A. It is determined whether or not the terminal ID acquired in step SP172 exists (SP179).

Obtaining a negative result in this determination means that the operation information subtype of the data transmitted from the data collection source terminal 2 to which the terminal ID acquired in step SP172 is assigned is the operation information subtype of the subtype name acquired in step SP177. Means not.

Thus, at this time, the residual data collection control unit 35 increases the value of the variable t by “1” (SP183), and thereafter, the priority column 42A (FIG. 10) on the priority management table 42 (FIG. 10). It is determined whether there is a record having the same priority as the variable t at that time (SP184). When the residual data collection control unit 35 obtains a positive result in this determination, it returns to step SP177, and thereafter obtains a positive result at step SP179 or obtains a negative result at step SP184 until it obtains a negative result at step SP177 to step SP179-step SP182. And the process of step SP184 is repeated.

On the other hand, obtaining an affirmative result in the determination of step SP179 is that the operation information subtype of the data transmitted from the data collection source terminal 2 to which the terminal ID acquired in step SP172 is assigned is the sub acquired in step SP177. It means that it is the operation information sub-type of the type name. Thus, at this time, the residual data collection control unit 35 acquires the operation information determination standard corresponding to the sub type name acquired in step SP172 from the operation information determination reference file 50 (FIG. 15) (SP180).

Subsequently, the residual data collection control unit 35 acquires the data collection method of the operation information sub type with the priority “t” derived by the residual data collection method derivation process described above with reference to FIG. 23 (SP181). Specifically, the residual data collection control unit 35 transmits the transmission interruption time of the record having the priority “t” stored in the priority column 46A (FIG. 14) from the residual data collection method management table 46 (FIG. 14). The transmission interruption time stored in the column 46C (FIG. 14), the transmission interval stored in the transmission interval column 46D (FIG. 14) of the record, and the transmission data number column 46E (FIG. 14) of the record The number of data items is acquired as the data collection method of the operation information sub type.

Next, the residual data collection control unit 35 registers the data collection method of the operation information sub type with the priority “t” for the residual data collection command generated in step SP173 (SP182). Specifically, the residual data collection control unit 35 sets the operation information sub-type having the priority “t” as the data type to be targeted at each corresponding location in the residual data collection command, and sets the transmission interruption time as the transmission interruption time. The transmission interruption time acquired in step SP181 is set, and the value obtained by dividing the number of data acquired in step SP181 by the transmission interval acquired in step SP181 is set as the transmission frequency.

Then, the residual data collection controller 35 thereafter processes step SP183 and subsequent steps in the same manner as described above.

Eventually, when the residual data collection control unit 35 obtains a positive result in step SP184 by completing the processing of step SP177 to step SP182 for the data collection source terminal 2 of the terminal ID acquired in step SP172, the value of the variable n is set. It is incremented by “1” (SP185), and it is determined whether or not the record of the “n” line exists in the terminal information table 40 (FIG. 8) (SP186).

Then, when the residual data collection control unit 35 obtains a negative result in this determination, it returns to step SP172, and thereafter repeats the processing of step SP172 to step SP186 until a positive result is obtained at step SP186.

When the residual data collection control unit 35 finally generates a residual data collection command for all the data collection source terminals 2 whose terminal IDs are registered in the terminal information table 40 and obtains a positive result in step SP186, the residual data collection control unit 35 The data collection command generation process is terminated.

Each residual data collection command generated in this way is then transmitted to the corresponding data collection source terminal 2 by the command delivery unit 36 (FIG. 1) as described above. Each data collection source terminal 2 that has received the residual data collection command stores the alert information and the corresponding operation information data in the order, timing, and frequency according to the collection method specified in the residual data collection command. It is transmitted to the collection / delivery server 4.

In this case, when the first residual data number acquisition processing method described above with reference to FIG. 21 is applied as the residual data number acquisition processing method in step SP81 of the residual data collection processing described above with reference to FIG. Since the terminal 2 has already obtained information on the alert information determination criteria and the operation information determination criteria for each operation information subtype (see step SP104 to step SP106 in FIG. 21), the terminal 2 of the main storage device 12 (FIG. 1) It is possible to determine which data among the data remaining in the buffer area 13 is alert information and which data is operation information. Therefore, in this case, each data collection source terminal 2 can transmit alert information and corresponding operation information to the data collection / delivery server 4 in the order, timing and frequency specified in the received residual data collection command.

However, when the second residual data number acquisition processing method described above with reference to FIG. 22 is applied as the residual data number acquisition processing method, each data collection source terminal 2 receives the alert information determination standard and each operation information sub The information of the type of operation information determination criterion is not obtained from the data collection / delivery server 4, and therefore, of the data remaining in the buffer area 13 of the main storage device 12, which data is alert information and which data is operation information Cannot be determined.

For this reason, when the second residual data number acquisition processing method is applied as the residual data number acquisition processing method, the above alert information determination criteria and the corresponding operation information subtype are previously applied to each data collection source terminal 2. Of the above-mentioned alert information and the above-mentioned alert information determination criteria and the corresponding operation information sub-type operation information determination criteria to each data collection source terminal 2 at an appropriate timing from the data collection / delivery server 4 Need to be notified.

In this case, for example, a residual data number acquisition processing method for notifying the data collection source terminal 2 of the above-described alert information determination criteria and the corresponding operation information sub-type operation information determination criteria from the data collection / delivery server 4 at an appropriate timing. When the second residual data number acquisition processing method is applied as described above, the command delivery unit 36 collects residual data in the processing from step SP102 to step SP108 of the first residual data number acquisition processing described above with reference to FIG. If the command is executed before or immediately after being sent to the data collection source terminal 2 or the information on the alert information judgment standard and the corresponding operation information sub-type operation information judgment standard is stored in the residual data collection command good.

(1-4) Effects of this Embodiment As described above, in the data collection and delivery system 1 of this embodiment, the data collection and delivery server 4 periodically determines the priority of each current data type as viewed from the service user. After data recovery and failure recovery, the uncollected residual data remaining in the data collection source tenant 3 at that time, the priority of each current data type viewed from the service user, and the current data collection and distribution Each data collection source terminal 2 is controlled to collect data at an appropriate order and timing according to the state of the server 4 and at an appropriate frequency.

Therefore, according to the data collection / distribution system 1, the data collection / distribution server 4 gives priority to the data of the data type having a higher priority when viewed from the service user after the failure recovery and at that time. Data can be collected at a frequency according to the data processing capability of the data collection / delivery server 4. In this way, according to the data collection and distribution system 1, it is possible to reduce the adverse effect on the service user due to such a failure and improve the reliability of the data collection and distribution server 4.

(2) Second Embodiment (2-1) Configuration of Data Collection / Distribution System According to This Embodiment FIG. 25, in which parts corresponding to those in FIG. A data collection and distribution system 70 is shown. The data collection / distribution system 70 is provided with one or a plurality of data storage servers 71, and each data collection source terminal 2 in the data collection source tenant 3 has a data storage server 71 predetermined for the data collection source terminal 2. Is different from the data collection and delivery system 1 (FIG. 1) according to the first embodiment.

The data storage server 71 is a server device having a function of temporarily storing data transmitted from each data collection source terminal 2 of the data collection source tenant 3, and the data collection source tenant via the second network 72. 3 is connected to each data collection source terminal 2 in the network 3, and is connected to a data collection / distribution server 74 via a third network 73.

The data storage server 71 includes a CPU 80 and a main storage device 81. The CPU 80 is a processor that controls operation of the entire data storage server 71. The main storage device 81 is composed of, for example, a semiconductor memory. The main storage device 81 is provided with a buffer area 82 for temporarily holding data transmitted from each data collection source terminal 2 and also stores a data storage unit 83. The data storage unit 83 stores the data transmitted from each data collection source terminal 2 in the buffer area 82, while collecting the data stored in the buffer area 82 in response to a request from the data collection / delivery server 74. This is a program having a function of transmitting to the server 74.

The data collection / distribution server 74 is not provided with a buffer area for temporarily storing data from the data collection source terminal 2 in the main storage device 21, and the data collection / delivery server 74 stores the data from the data collection source terminal 2 in the main storage device 21. It is different from the data collection / delivery server 4 of the first embodiment in that it is acquired from the data storage server 71 by the data processing unit 90 stored in the first embodiment.

FIG. 26 shows a case where data transmitted from the data collection source terminal 2 in the data collection source tenant 3 to the data storage server 71 corresponds to the operation information in the data collection and distribution system 70 of the present embodiment. The flow of a series of processing is shown. Since the processing of step SP190 to step SP192 is the same as that of step SP1 to step SP3 of the first embodiment described above with reference to FIG. 3, description thereof is omitted here.

However, in the case of this embodiment, the data communication control unit 14 of the data collection source terminal 2 transmits the data stored in the buffer area 13 to a predetermined data storage server 71, and the data is stored in the data storage server 71 is stored in the buffer area 82 of the main storage device 21 (FIG. 25).

Thereafter, the data stored in the buffer area 82 of the data storage server 71 is periodically read out by the data processing unit 90 of the data collection / delivery server 74 (SP193). The data read from the buffer area 82 of the data storage server 71 by the data processing unit 90 of the data collection / distribution server 74 is thereafter deleted from the buffer area 82.

The data read by the data processing unit 90 is based on the operation information determination criteria stored in the operation information determination criteria file 50 and the alert information determination criteria stored in the alert information determination criteria file 51. (Alert information or operation information sub-type) is determined (SP194).

If it is determined by this determination that the data corresponds to the operation information (that is, the measurement value stored in the data is determined to be within the normal range as the measurement value of the corresponding operation information sub-type) The data is registered in the accumulation table 41 (SP195).

Thereafter, when a data search / delivery request specifying conditions such as operation information sub-type and storage date / time is given from the external server 7 to the data collection / delivery server 74 (SP196), the conditions specified in the data search / delivery request are satisfied. Data search is performed on the accumulation table 41 by the data processing unit 90 (SP197).

The data detected by this search is distributed to the external server 7 that is the transmission source of the data search distribution request (SP198), the contents of the data, the URL of the external server 7 that is the distribution destination, the distribution date and time, etc. Is written in the search distribution log file 48 as a search distribution log (SP195).

FIG. 27 shows a flow of a series of processing when the data read from the data storage server 71 corresponds to alert information. In this case, the processing from step SP200 to step SP204 is the same as the processing from step SP190 to step SP194 described above with reference to FIG. 26, and thus description thereof is omitted here.

If it is determined in step SP203 that the data read from the buffer area 82 of the data storage server 71 corresponds to the alert information, the data processing unit 90 immediately distributes the data to the external server 7 (SP205). Further, history information such as the contents of the data, the URL of the external server of the distribution destination, and the distribution date and time is written as a high-speed distribution log in the high-speed distribution log file 49 (SP206).

On the other hand, FIG. 28 shows the data collection source terminal 2, the data storage server 71, the data collection / distribution server 74, and the external server 7 when a failure occurs in the data collection / distribution server 74 so that the data processing necessary for providing the service cannot be performed. The state of is shown. FIG. 28 shows an example in which the failure that occurred in the data collection / delivery server 74 is a failure in which data cannot be acquired from the data storage server 71 and no data search / delivery request from the external server 7 can be received. ing.

The data from each data collection source terminal 2 is sequentially accumulated in the data storage server 71 even during the occurrence of a failure in the data collection and distribution server 74 (SP210). As described above, data stored in the buffer area 82 of the data storage server 71 includes data of each data type (alert information and each operation information sub-type). Therefore, the data collection / distribution server 74 prioritizes data of a data type having a high priority when viewed from the service user from among a large amount of data stored in the buffer area 82 of the data storage server 71 after the failure is recovered. Need to be acquired.

Therefore, in the data collection / distribution system 70 of the present embodiment, when the data collection / distribution server 74 recovers from the failure as described above with reference to FIG. Each data type in an appropriate order, timing and frequency according to the current state of the data collection / distribution server 74 and the number of residual data for each data type remaining in the buffer area 82 of the data storage server 71 at that time A second residual data collection function for collecting the data from the data storage server 71 is mounted on the data collection and distribution server.

In practice, in the case of the data collection / distribution system 70, the data collection / distribution server 74 periodically determines the priority of each current operation information sub-type viewed from the service user in the same manner as in the first embodiment. Detect and manage.

In addition, when the data collection / delivery server 4 recovers from the failure as described above with reference to FIG. 28, the data collection / delivery server 4 detects the current data processing capability with respect to the data processing capability during normal operation in the same manner as in the first embodiment. .

Further, the data collection / delivery server 74 obtains the data storage server 71 by, for example, counting the number of data of each data type remaining in the buffer area 82 of the data storage server 71 by itself.

The data collection / distribution server 74 determines the priority of each operation information sub-type managed at that time, the current data processing capability, and the number of data items for each data type remaining in the data storage server 71. Based on the data type, a data collection method (residual data collection method) for deriving the data in what order, timing and frequency is derived. At this time, the data collection / distribution server 74 derives a residual data collection method for collecting alert information earlier (first) than the operation information.

The data collection / distribution server 74 obtains data of each data type from the data storage server 71 in the order, timing and frequency according to the residual data collection method derived in this way.

As a function for realizing the second residual data collection function according to the present embodiment as described above, the main storage device 21 of the data collection / distribution server 74 includes a data processing unit 90, a failure, as shown in FIG. A detection unit 31, an operation information priority detection unit 91, a server state detection unit 33, a residual data number acquisition unit 92, a residual data collection control unit 93, and a command delivery unit 36 are stored. Among these, the failure detection unit 31, the server state detection unit 33, and the command delivery unit 36 are programs having the functions described above with respect to the first embodiment, and thus description thereof is omitted here.

The data processing unit 90 periodically receives data from each data collection source terminal 2 accumulated in the buffer area 82 of the main storage device 81 of the data storage server 71 from the data storage server 71 via the third network 73. It is a program having a function to acquire.

Further, the data processing unit 90 stores the data type of the data acquired from the data storage server 71 in the operation information determination standard file 50 (FIG. 15) stored in the operation information determination standard file 50 (FIG. 15) and the alert information determination standard file 51 (FIG. 16). Judgment is made based on the stored alert information judgment criteria.

When the data processing unit 90 determines that the data type of the data is operation information, the data processing unit 90 registers the data in the accumulation table 41. The data processing unit 90 then reads out data satisfying the condition from the storage table 41 and transmits it to the external server 7 in response to a data search / distribution request from the external server 7, and the history information (search / distribution log). Is stored in the search distribution log file 48 (FIG. 25). When the data processing unit 30 determines that the data is alert information in the determination process, the data processing unit 30 distributes the alert information to the external server 7 at a high speed, and the history information (high-speed distribution log) Store in file 49 (FIG. 25).

The operation information priority detection unit 91 is a program having a function of periodically detecting the priority of each current data type viewed from the service user based on the search distribution log stored in the search distribution log file 48. It is. The operating information priority detection unit 91 updates a priority management table 100 described later based on the detection result.

Further, the remaining data number acquisition unit 92 is a program having a function of acquiring the number of data of each data type remaining in the buffer area 82 of the data storage server 71 when the server recovers from a failure. . The residual data count acquisition unit 92 registers and manages the acquired data count for each data type in the residual data count management table 45.

The residual data collection control unit 93 determines the current priority for each data type detected by the operation information priority detection unit 32 and the own server detected by the server state detection unit 33 when the own server recovers from a failure. The residual data collection method described above is derived on the basis of the current state of the data and the number of residual data items for each data type remaining in the buffer area 82 of the data storage server 71 acquired by the residual data number acquisition unit 92. It is a program having a function. The residual data collection control unit 93 registers and manages the derived residual data collection method in a residual data collection method management table 102 described later. The residual data collection control unit 93 acquires data of each data type from the data storage server 71 in the order, timing, and frequency according to the residual data collection method registered in the residual data collection method management table 102.

On the other hand, the auxiliary storage device 22 (FIG. 25) of the data collection and distribution server 74 has a terminal information table 40, an accumulation table 41, a priority as means for realizing the second residual data collection function of the present embodiment. Degree management table 100, data acquisition destination server management table 101, server state management table 44, residual data number management table 45, residual data collection method management table 102, specification information file 47, search distribution log file 48, high-speed distribution log file 49 The operation information determination standard file 50 and the alert information determination standard file 51 are stored.

Among these, the terminal information table 40, the accumulation table 41, the server state management table 44, the residual data number management table 45, the specification information file 47, the search distribution log file 48, the high-speed distribution log file 49, the operation information determination reference file 50, and the alert information The determination reference file 51 is information having the configuration as described above with respect to the first embodiment, and thus description thereof is omitted here.

The priority management table 100 is a table used for managing the priority for each operation information sub type, and as shown in FIG. 29, the priority column 100A, the operation information sub type column 100B, the collected data acquisition destination A server column 100C and a search distribution execution number column 100D are provided. In the priority management table 100, one row corresponds to one priority.

In the priority column 100A, the operation information sub-type column 100B, and the search delivery execution count column 100D, the priority column 42A and the operation information sub-type of the priority management table 42 according to the first embodiment described above with reference to FIG. Information similar to the information stored in the corresponding columns in the column 42B and the search distribution execution count column 42D is stored. The collected data acquisition destination server column 100C stores the server name of the data storage server 71 in which the corresponding operation information sub-type data is stored. FIG. 29 shows an example in which all data collection source terminals 2 transmit data to the data storage server 71 having the server name “DATA_Server1”.

The data acquisition server management table 101 is a table used for managing the name (cache name) of the buffer area 82 in which data from the corresponding data collection source terminal 2 in each data storage server 71 is stored. The server name (acquisition destination server name) of each data storage server 71 and the name of the buffer area 82 (acquisition destination cache name) for storing data from the data collection source terminal 2 in the corresponding data storage server 71 are stored in association with each other. Is done.

On the other hand, the residual data collection method management table 102 is a table used for managing the above-described residual data collection method derived by the residual data collection control unit 93 (FIG. 25). As shown in FIG. A priority column 102A, a processing time column 102B, an acquisition interruption time column 102C, an acquisition interval column 102D, and an acquisition data number column 102E are configured.

The priority column 102A, the processing time column 102B, the acquisition interruption time column 102C, the acquisition interval column 102D, and the acquisition data number column 102E are respectively stored in the residual data collection method management of the first embodiment described above with reference to FIG. Information similar to the information stored in the corresponding columns of the priority column 46A, the processing time column 46B, the transmission interruption time column 46C, the transmission interval column 46D, and the transmission data number column 46E of the table 46 is stored.

(2-2) Various Processes Related to Second Residual Data Collection Function According to this Embodiment (2-2-1) Operation Information Priority Allocation Processing The operation information priority detection unit 91 according to this embodiment is shown in FIG. According to the same processing procedure as the operation information priority assignment process of the first embodiment described above and the sub type priority registration process of the first embodiment described above with reference to FIG. The priority of the data type is detected, and the priority management table 100 is updated based on the detection result.

However, the operation information priority detection unit 91 according to the present embodiment acquires the server name of the data storage server 71 in which the data of the operation information sub type obtained in step SP54 in step SP57 of FIG. 17 is stored. The server name acquired from the destination server management table 101 and registered in the priority management table 100 (FIG. 29) in step SP70 of FIG. Stored in the collected data acquisition destination server column 100C of the management table 100.

(2-2-2) Flow of Residual Data Acquisition Processing FIG. 31 shows a flow of residual data acquisition processing executed in the data collection / distribution server 74 after a failure occurs in the data collection / distribution server 74 and recovery from the failure. Indicates.

In the data collection / distribution server 74, after recovery from such a failure, first, the server state detection unit 33 (FIG. 1) executes server state detection processing for detecting the current server state of the data collection / distribution server 4 (SP220). ). The specific processing content of this server state detection process is the same as the server state detection process according to the first embodiment described above with reference to FIG.

Subsequently, the residual data number acquisition unit 92 (FIG. 25) executes a residual data number acquisition process for acquiring the number of residual data remaining in the buffer area 82 of the data storage server 71 for each data type (SP221). ).

Next, the residual data collection control unit 93 (FIG. 25) determines the priority of each current operation information subtype viewed from the service user registered in the priority management table 100 (FIG. 29), and step SP221. The order, timing, and frequency for each data type for collecting these residual data based on the number of data items for each data type (residual data) remaining in the buffer area 82 of the data storage server 71 acquired in FIG. The residual data collection method deriving process for deriving the residual data collection method that defines the above is executed (SP222).

Then, the residual data collection controller 93 executes a residual data acquisition process for acquiring residual data from the data storage server 71 at the order, timing and frequency for each data type according to the residual data collection method derived at step SP222 (SP223). ).

As described above, all data of each data type accumulated in the data storage server 71 during the failure of the data collection / delivery server 74 is collected in the data collection / delivery server 74, and necessary information of these data is stored in the accumulation table 41 (FIG. 25). ). Thereafter, the residual data collection process ends.

(2-2-3) Residual Data Number Acquisition Process FIG. 32 shows an example of a residual data number acquisition process executed by the residual data number acquisition unit 92 in step SP221 of the residual data collection process described above with reference to FIG. The residual data number acquisition processing method shown in FIG. 32 is performed by the residual data number acquisition unit 92 determining the data type of the residual data existing on the buffer area 82 of the data storage server 71 one by one. In this method, the number of remaining data items in the storage server 71 is acquired for each data type.

In practice, when the method shown in this figure is applied as the residual data number acquisition process, the residual data number acquisition unit 92 first determines whether or not a record exists in the residual data number management table 45 (FIG. 13). (SP230). When the residual data number acquisition unit 92 obtains a negative result in this determination, it proceeds to step SP232, and when it obtains a positive result, it deletes all the records existing on the residual data number management table 45 (SP231).

Subsequently, the residual data number acquisition unit 92 generates a variable [data type name] _C for each data type for counting the number of residual data with an initial value “0” (SP232). Note that “variable [data type name] _C” is a place where the type name of the corresponding data type (in this embodiment, alert information, pressure information, temperature information or vibration information) is “[data type name]”. A variable to be stored.

Next, the residual data number acquisition unit 92 acquires record information (acquisition destination server name and acquisition destination cache name) of each record from the data acquisition destination server management table 101 (SP233). In the present embodiment, it is assumed that there is only one record registered in the data acquisition destination server management table 233 as described above.

Thereafter, the remaining data number acquisition unit 92 sets the variable n to the initial value “1” (SP234), and further in step SP233 of the data storage server 71 to which the acquisition destination server name acquired in step SP233 is assigned. The contents of the n-th residual data (value of each operation information sub type stored in the residual data) are acquired from the buffer area 82 to which the acquired acquisition destination cache name is assigned (SP235).

Subsequently, the remaining data number acquisition unit 92 operates for each alert information stored in the alert information determination reference file 51 (FIG. 16) and each operation information subtype stored in the operation information determination reference file 50 (FIG. 15). While sequentially reading out the information determination criteria, the data type of the residual data acquired at step SP is determined by sequentially comparing these with the content of the nth residual data acquired at step SP235 (SP236). ). Then, the residual data number acquisition unit 92 increases (increments) the value of the variable [data type name] _C corresponding to the data type of the residual data detected by this comparison by “1” (SP237).

Next, the residual data number acquisition unit 92 increases the value of the variable n by “1” (SP238), and then in step SP233 of the data storage server 71 to which the acquisition destination server name acquired in step SP233 is assigned. It is determined whether or not there is n-th residual data in the buffer area 82 to which the acquired acquisition destination cache name is assigned (SP239). The residual data number obtaining unit 92 returns to step SP235 when obtaining a positive result in this determination, and thereafter repeats the processing of step SP235 to step SP239 until obtaining a positive result at step SP239.

When the residual data number acquisition unit 92 finally obtains a positive result in step SP239 by completing the processing of step SP236 to step SP237 for all residual data remaining in the buffer area 82, each variable [ The data type name] _C value is registered in the residual data number management table 45 (FIG. 13) (SP240). Specifically, the residual data number acquisition unit 92 sets the value of each variable [data type name] _C in the location of “[data type name]” of the variable [data type name] _C in the residual data number management table 45, respectively. Is stored in the remaining data number column 45B (FIG. 13) of the record stored in the data type column 45A (FIG. 13).

Then, the residual data number acquisition unit 92 thereafter ends this residual data number acquisition processing.

As another example of the residual data number acquisition process executed in step SP221 of the residual data acquisition process described above with reference to FIG. 31, the residual data number acquisition unit 92 of the data collection / distribution server 74 has a unit time during normal operation. The number of data processing per (one minute) is calculated for each data type, and the number of data remaining in the buffer area 82 of the current data storage server 71 is estimated for each data type based on the calculation result. A method to do this is also possible. Since the processing content of the residual data number acquisition unit by this method is the same as the second residual data number processing described above with reference to FIG. 22, the description thereof is omitted here.

(2-2-4) Residual Data Collection Method Derivation Process FIG. 33 shows a residual data collection method derivation process executed by the residual data collection control unit 93 (FIG. 25) in step SP222 of the residual data collection process described above with reference to FIG. The specific processing procedure of is shown.

When the residual data collection control unit 93 starts the residual data collection method derivation process shown in FIG. 33, it first determines whether or not a record exists in the residual data collection method management table 102 (FIG. 30) (SP250). . If the residual data collection control unit 93 obtains a negative result in this determination, it proceeds to step SP252, and if it obtains a positive result, it deletes all the records existing in the residual data collection method management table 102 (SP251).

Subsequently, the residual data collection control unit 93 acquires the current data processing capability of the server itself from the server state management table 44 (FIG. 12) (SP252). Specifically, the residual data collection control unit 35 stores the value (current number of data processing per minute) stored in the processing number column 44D (FIG. 12) of the row corresponding to the own server in the server state management table 44. To get.

Next, the residual data collection control unit 93 obtains the number of residual data of the alert information remaining in the buffer area 82 of the data storage server 71 at that time from the residual data number management table 45 (FIG. 13) (SP253). Specifically, the residual data collection control unit 93 includes the residual data count column 45B in the row corresponding to the alert information in the residual data count management table 45 (the row in which “ARERT” is stored in the data type column 45A (FIG. 13)). The value stored in (FIG. 13) is acquired.

Furthermore, the residual data collection controller 93 divides the number of residual data acquired in step SP253 by the current data processing capability of the local server acquired in step SP252, thereby remaining in the buffer area 82 of the data storage server 71. The time (processing time) required to process the data corresponding to all the alert information currently being processed with the data processing capability of the current server is calculated (SP254).

Then, the residual data collection controller 93 registers the alert information processing time and acquisition interruption time in the residual data collection method management table 102 (FIG. 30) (SP255). Specifically, the residual data collection control unit 93 secures an unused line on the residual data collection method management table 102, sets “0” in the priority column 102A (FIG. 30) of that line, and the processing time of that line. The processing time calculated in step SP254 is stored in the column 102B (FIG. 30), and “0” is stored in the acquisition interruption time column 102C (FIG. 30) of that row.

Subsequently, the residual data collection controller 93 sets the variable m to the initial value “1” (SP256). Further, the residual data collection control unit 93 acquires the number of remaining data items of the operation information sub type with the priority “m” remaining in the buffer area 82 of the data storage server 71 (SP257). Specifically, the residual data collection control unit 93 refers to the priority management table 100 (FIG. 29), and the operation information sub-record of the record whose value stored in the priority column 100A (FIG. 29) is “m”. The sub type name stored in the type column 100B (FIG. 29) is acquired, and among the records in the residual data number management table 45 (FIG. 13), the above-mentioned sub type name is stored in the data type name column 45A (FIG. 13). The value stored in the stored residual data number column 45B (FIG. 13) is acquired.

Next, the residual data collection control unit 93 divides the number of data acquired in step SP257 by the current data processing capability of the local server acquired in step SP252, thereby remaining in the buffer area 82 of the data storage server 71. The time (processing time) required to process all the data storing the operation information of the operation information subtype having the priority “m” with the current data processing capability of the own server is calculated (SP258).

Further, the residual data collection control unit 93 calculates the acquisition interruption time of data in which the operation information of the operation information subtype with the priority “m” is stored (SP259). Specifically, the residual data collection control unit 93 selects all the records registered in the residual data collection method management table 102 (FIG. 30) whose priority stored in the priority column 102A is less than “m”. The sum of the values stored in the record processing time column 102B is calculated as the acquisition interruption time of the data storing the operation information of the operation information subtype with the priority “m”.

The “acquisition interruption time” is the data type (alert information) having the highest priority when the data collection / distribution server 74 acquires data of each data type from the data storage server 71 in the order according to the priority. Indicates the time from the start of data acquisition until the start of data acquisition of the data type to which the corresponding priority is assigned.

Thereafter, the residual data collection control unit 93 secures an unused line on the residual data collection method management table 102 (FIG. 30), “m” in the priority column 102A of that line, and the processing time of that line. The processing time acquired in step SP258 is stored in the column 102B, and the acquisition interruption time acquired in step SP260 is stored in the acquisition interruption time column 102C of the row (SP260).

Thereafter, the residual data collection control unit 93 increases the value of the variable m by “1” (SP261), and then refers to the priority management table 100 (FIG. 29) to determine the value of the variable m at that time. It is determined whether or not there is an operation information sub type assigned as a priority (SP262). If the residual data collection controller 93 obtains a negative result in this determination, it returns to step SP257, and thereafter repeats the processing of step SP257 to step SP262 until a positive result is obtained in step SP262.

Through the processing of step SP257 to step SP262 described above, the data acquisition timing for each data type (alert information and each operation information subtype) stored in the buffer area 82 of the data storage server 71 is derived.

When the residual data collection control unit 93 finally obtains a positive result in step SP262 by deriving the acquisition timing for each data type, the number of alert information data items to be acquired from the data storage server 71 per minute (hereinafter, This is called the alert information acquisition frequency) (SP263). Specifically, the residual data collection control unit 93 sets the remaining data count of the alert information acquired in step SP253 to the processing time column 102B of the record corresponding to the alert information in the residual data collection method management table 102 (FIG. 30) (FIG. The alert information acquisition frequency is calculated by dividing by the value stored in 30).

The residual data collection control unit 93 stores the alert information acquisition frequency calculated in this way in the acquisition data number field 102E (FIG. 30) of the corresponding record in the residual data collection method management table 102, and the acquisition interval field. “1” is stored in 102D (FIG. 30) (SP264).

Subsequently, the residual data collection controller 93 sets the variable x to the initial value “1” (SP265). Further, the residual data collection control unit 93 acquires the sub type name of the operation information sub type whose priority is set to “x” in the priority management table 100 (FIG. 29) (SP266).

Next, the residual data collection control unit 93 stores data of residual data in which the operation information of the operation information subtype acquired in step SP266 and remaining in the buffer area 82 of the data storage server 71 is stored. The number of cases is acquired from the residual data number management table 45 (FIG. 13) (SP267).

Then, the residual data collection control unit 93 stores the number of residual data items in which the operation information of the operation information sub-type with the priority “x” to be acquired from the data storage server 71 in one minute (hereinafter, this is the priority). Is called the operation information acquisition frequency of the operation information sub-type with “x” (SP268). Specifically, the residual data collection control unit 93 determines the number of residual data items of the operation information sub-type with the priority “x” acquired in step SP267 as the priority of the residual data collection method management table 102 (FIG. 30). Is obtained by dividing by the value stored in the processing time column 102B (FIG. 30) of the record corresponding to the operation information sub type of “x”, to obtain the operation information of the operation information sub type of the priority “x”. Calculate the frequency.

Then, the residual data collection control unit 93 sets the frequency of operation information of the operation information sub-type having the priority “x” calculated in this way as the corresponding record in the residual data collection method management table 102 (FIG. 30). Stored in the acquisition data number column 102E (FIG. 30) and “1” in the acquisition interval column 102D (FIG. 30) (SP269).

Thereafter, the residual data collection control unit 93 increases the value of the variable x by “1” (SP270), and then refers to the priority management table 100 (FIG. 29) to determine the value of the variable x at that time. It is determined whether or not there is an operation information sub type assigned as priority (SP271). The residual data collection controller 93 returns to step SP266 when a negative result is obtained in this determination, and thereafter repeats the processing of step SP266 to step SP271 until a positive result is obtained at step SP271.

Through the processes in steps SP266 to SP271 described above, the operation information acquisition frequency for each operation information subtype is derived.

When the residual data collection control unit 93 finally obtains a positive result in step SP271 by completing the processing of step SP267 to step SP269 for all priorities assigned to the operation information subtype, the residual data collection method is derived. The process ends.

(2-2-5) Residual Data Acquisition Processing FIG. 34 shows a specific processing procedure of the residual data acquisition processing executed by the residual data collection control unit 93 in step SP223 of the residual data collection processing described above with reference to FIG. Show.

When the residual data collection control unit 93 ends the residual data collection method derivation process described above with reference to FIG. 33, the residual data collection control unit 93 starts the residual data acquisition process shown in FIG. The alert information determination criteria that have been set are acquired (SP280).

Subsequently, the residual data collection control unit 93 acquires the alert information collection method derived by the residual data collection method derivation process described above with reference to FIG. 33 (SP281). Specifically, the residual data collection control unit 93 obtains the record interruption time of the record whose priority stored in the priority column 102A (FIG. 30) is “0” from the residual data collection method management table 102 (FIG. 30). The acquisition interruption time stored in the column 102C (FIG. 30), the acquisition interval stored in the acquisition interval column 102D (FIG. 30) of the record, and the acquisition data number column 102E (FIG. 30) of the record The number of data is acquired as a method for collecting residual data of alert information.

Then, the residual data collection control unit 93 waits for the alert interruption time of the alert information acquired at step SP281 (SP282), and thereafter, the data processing unit 90 acquires all the residual data of the alert information from the data storage server 71. An instruction is given to (FIG. 25) (SP283). As a result, the data processing unit 90 that has received such an instruction acquires all the residual data of the alert information from the buffer area 82 of the data storage server 71, and the necessary information of these residual data is stored in the accumulation table 41 (FIG. 9). Each is registered.

Subsequently, the residual data collection controller 93 sets the variable “t” to the initial value “1” (SP284) and stores it in the priority column 100A (FIG. 29) from the priority management table 100 (FIG. 29). The sub type name stored in the operation information sub type column 100B (FIG. 29) of the record having the priority “t” is acquired (SP285).

Next, the residual data collection control unit 93 acquires the operation information determination standard corresponding to the sub type name acquired in step SP285 from the operation information determination reference file 50 (FIG. 15) (SP286).

Further, the residual data collection control unit 93 acquires the data collection method of the operation information sub type with the priority “t” derived by the residual data collection method derivation process described above with reference to FIG. 33 (SP287). Specifically, the residual data collection control unit 93 obtains the record interruption time of the record whose priority stored in the priority column 102A (FIG. 30) is “t” from the residual data collection method management table 102 (FIG. 30). The acquisition interruption time stored in the column 102C (FIG. 30), the acquisition interval stored in the acquisition interval column 102D (FIG. 30) of the record, and the acquisition data number column 102E (FIG. 30) of the record The number of data items is acquired as the data collection method of the operation information sub type.

Subsequently, the residual data collection control unit 93 waits until the acquisition interruption time of the corresponding operation information sub type acquired in step SP287 elapses (SP288), and thereafter, the data storage server 71 sets the operation information sub type. An instruction is given to the data processing unit 90 to acquire the residual data in which the operation information is stored (SP289). As a result, the data processing unit 90 that has received such an instruction acquires all the residual data in which the operation information of the operation information sub type is stored from the buffer area 82 of the data storage server 71, and the necessary information of these residual data. Are registered in the accumulation table 41 (FIG. 9).

Next, the residual data collection controller 93 increases the value of the variable t by “1” (SP290), and thereafter, there is an operation information sub-type for which residual data has not yet been acquired from the data storage server 71. Is determined (SP291). If the residual data collection control unit 93 obtains a positive result in this determination, it returns to step SP285, and thereafter repeats the processing of step SP285 to step SP291 until a negative result is obtained in step SP291.

When the residual data collection control unit 93 finally obtains a positive result in step SP291 by completing the acquisition of all the residual data in which the operation information of any operation information subtype is stored from the data storage server 71, this residual data The data acquisition process is terminated.

(2-3) Effects of this Embodiment As described above, in the data collection and delivery system 1 of this embodiment, the data collection and delivery server 74 periodically determines the priority of each data type viewed from the service user. After the failure is recovered and the failure is recovered, the uncollected residual data remaining in the data storage server 71 at that time, the current priority of each data type viewed from the service user, and the current data collection and distribution server 74 The data is collected at an appropriate order and timing according to the state, and at an appropriate frequency.

Therefore, according to the data collection / distribution system 70, after the failure recovery, the data collection / distribution server 4 gives priority to the data of the data type having a higher priority when viewed from the service user at that time. Data can be collected at a frequency according to the data processing capability of the data collection / delivery server 4. Thus, according to the data collection / distribution system 70, similarly to the data collection / distribution system 1 of the first embodiment, the adverse effect on the service user due to the failure is reduced, and the data collection / distribution server The reliability of 74 can be improved.

(3) Other Embodiments In the first and second embodiments described above, the sensor 10 (FIGS. 1 and 25) is mounted as a terminal that transmits data to the data collection and

distribution servers

4 and 74. The data collection source terminal 2 is applied, and the present invention is applied to the data collection and

distribution systems

1 and 70 in which the data collection and

distribution servers

4 and 74 collect measurement values representing the state of the measurement target output from the sensor 10 as data. However, the present invention is not limited to this. In short, one or a plurality of terminals, an external device used by a service user, and data of a plurality of data types transmitted from the terminals are stored. In response to a search / delivery request from an external device, the data collection / distribution device configured to distribute data corresponding to the data type from the stored data to the external device. The present invention can be widely applied to data collection / distribution systems having various configurations.

In the first and second embodiments described above, the specification information file 47 (FIGS. 1 and 25) in which information representing the data processing capability of the data collection and

distribution servers

4 and 74 during normal operation is stored is stored in advance. The server status detection unit 33 (FIGS. 1 and 25) uses the data processing capability of the data collection /

delivery servers

4 and 74 during normal operation stored in the specification information file 47 to recover the failure. Although the case where the state of the data collection /

delivery servers

4 and 74 is detected has been described, the present invention is not limited to this, and the high-speed delivery log file 49 (FIGS. 1 and 25) and the accumulation table 41 (FIGS. 1 and 25) Based on the data processing contents during normal operation registered in 25), the server status detection unit 33 calculates the data processing capability of the data collection and

distribution servers

4 and 74 during normal operation by calculation. It may be.

INDUSTRIAL APPLICABILITY The present invention can be widely applied to data collection / distribution apparatuses having various configurations that collect and accumulate data from a terminal or the like and provide the accumulated data to the service user in response to a request from the service user. .

DESCRIPTION OF

SYMBOLS

1,70 ... Data collection delivery system, 2 ... Data collection origin terminal (terminal), 3 ... Data collection origin tenant, 4,74 ... Data collection delivery server, 7 ... External server (external device), 10 ... Sensor, 20 ... CPU, 90, 90 ... Data processing section, 31 ... Failure detection section, 32, 91 ... Operation information priority detection section, 33 ... Server status detection section, 34, 92 ... Residual data count acquisition unit, 35, 93 ... Residual data collection control unit, 36 ... Command delivery unit, 37, 82 ... Buffer area, 40 ... Terminal information table, 41 ... Accumulation table, 42, 100 ... Priority management table 43 …… Terminal list group 43A …… Terminal list 44 …… Server status management table 45 …… Residual data number management table 46,102 …… Residual data collection method management table , 47 …… Spec information file, 48 …… Search delivery log file, 49 …… High speed delivery log file, 50 …… Operation information judgment reference file, 51 …… Alert information judgment reference file, 71 …… Data storage server, 101: Data acquisition destination server management table.

Claims

Stores data of a plurality of data types transmitted from one or a plurality of terminals, and distributes data of the corresponding data types from the stored data to the external device in response to a search distribution request from the external device. In the data collection device
A priority detection unit that detects the priority of each current data type based on the past search distribution request;
A data number acquisition unit for acquiring the data number of the uncollected data for each data type;
A server state detection unit for detecting the current data processing capability of the own device;
The priority of each data type currently detected by the priority detection unit, the number of data acquired by the data number acquisition unit after recovery from a failure, and the server status detection unit after recovery from the failure A data collection control unit for deriving a collection method of the uncollected data in consideration of the current data processing capability of the own device detected by the device and collecting the uncollected data according to the derived collection method; A data collection device comprising:
The data collection control unit
The order and timing for each data type for which the data is to be collected and the number of the data to be collected per unit time are derived as the collection method for the uncollected data. The data collection device described in 1.
The data collection control unit
By generating a command for each terminal designating the transmission timing of the data according to the derived collection method and the transmission frequency of the data, and transmitting each generated command to the corresponding terminal. The data collection device according to claim 2, wherein uncollected data is collected from each terminal according to the collection method.
Each of the terminals transmits the data to a data storage server,
The data storage server temporarily holds the data transmitted from each terminal,
The data collection control unit
The data collection device according to claim 2, wherein the uncollected data is collected from the data storage server in accordance with the derived collection method.
The priority detection unit
2. The data collection according to claim 1, wherein the priority of each of the current data types is detected based on the number of search delivery requests for the data for each of the data types in a most recent predetermined period. apparatus.
The data count acquisition unit
Calculate the average value of the number of acquisitions for each data type per unit time before the occurrence of the failure, the calculated average value of the number of acquisitions for each data type per unit time, and the current value after the occurrence of the failure The data collection device according to claim 1, wherein the data collection device is obtained by estimating the number of data items of the uncollected data for each data type based on an elapsed time until.
Stores data of a plurality of data types transmitted from one or a plurality of terminals, and distributes data of the corresponding data types from the stored data to the external device in response to a search distribution request from the external device. In the data collection method executed by the data collection device
A first step of detecting a priority of each current data type based on the past search distribution request;
A second step of obtaining the number of uncollected data for each data type and detecting the current data processing capability of the device;
The current priority of each data type, the number of uncollected data for each data type acquired after recovery from a failure, and the current data processing capability of the own device detected after recovery from the failure A data collection method comprising: a third step of deriving a collection method of the uncollected data considered and collecting the uncollected data according to the derived collection method.
In the second step,
8. The method of collecting the uncollected data includes deriving an order and timing for each data type from which the data is to be collected, and a number of the data to be collected per unit time. The data collection method described in 1.
In the third step,
By generating a command for each terminal designating the transmission timing of the data according to the derived collection method and the transmission frequency of the data, and transmitting each generated command to the corresponding terminal. The data collection method according to claim 8, wherein the uncollected data is collected from each terminal according to the collection method.
Each of the terminals transmits the data to a data storage server,
The data storage server temporarily holds the data transmitted from each terminal,
In the third step,
The data collection method according to claim 8, wherein the uncollected data is collected from the data storage server in accordance with the derived collection method.
In the first step,
The data collection according to claim 7, wherein the priority of each of the current data types is detected based on the number of search delivery requests for the data for each of the data types in a most recent predetermined period. Method.
In the second step,
Calculate the average value of the number of acquisitions for each data type per unit time before the occurrence of the failure, the calculated average value of the number of acquisitions for each data type per unit time, and the current value after the occurrence of the failure The data collection method according to claim 7, wherein the data collection method is obtained by estimating the number of data pieces of the uncollected data for each data type based on the elapsed time until.