JP2017069661A - Data collection apparatus, data collection system, and data collection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data collection apparatus capable of completing information collection from a plurality of access points within one cycle.SOLUTION: The data collection apparatus collects data which are stored in a plurality of objective apparatuses in a manner of plural processing which is capable of making parallel access to the plurality of objective apparatuses. The data collection apparatus includes an acquisition section that, with respective objective apparatuses, acquires the data by allotting an objective apparatus which makes accesses to the a plurality of processes, in an order of longer estimated data collection time.SELECTED DRAWING: Figure 9

Description

  The present invention relates to a data collection device, a data collection system, and a data collection method.

  With the spread of wireless terminals such as smartphones and tablet terminals, services that enable wireless terminal users to obtain information by connecting the wireless terminals to a network such as the Internet are increasing.

  For example, wireless LAN (Local Area Network) access points (hereinafter referred to as “AP”) are arranged at various locations such as stations, airports, office buildings, shops, and outdoors so that services can be provided throughout Japan. As a result, the number of APs is increasing rapidly.

  The AP accumulates data useful for association analysis, such as data indicating the usage status of wireless terminals (for example, connection information for each terminal). Conventionally, a collection server that collects data from each AP and a collection management server that manages data collected from each AP are arranged in a network. Each collection server polls each of a plurality of APs and collects data accumulated in the APs. As the number of APs increases, the number of collection servers also increases.

JP-A-9-321758 Japanese Patent Laid-Open No. 10-41877 Special table 2008-512047 gazette

  When collecting information stored in a plurality of APs by a plurality of collection servers, there are the following two methods for determining the AP from which each collection server collects data (AP allocation method for each collection server): .

  One is a method in which a plurality of APs are fixedly assigned to each collection server in advance and each collection server collects data from all of the plurality of assigned APs within one period in which data is collected from the APs (“ Referred to as "first method"). The other is a method in which each collection server selects an AP that has not been collected within a certain period (referred to as a “second method”).

  The information amount of the collected data varies depending on the number of people traveling around the AP and the usage status of the wireless terminal. In the first method, when the amount of data collected from a certain AP swells due to temporary congestion in the AP placement area (increase in AP users), the time required for data collection is one cycle. There was a case where the time length was exceeded.

  Usually, the processing capacity of the collection server is determined so that it can be processed with a certain margin with respect to the peak processing load. For this reason, even if the temporary congestion of the AP placement area as described above rarely occurs, the collection server facilities and the like are set so that the data collection at the time of congestion is completed within one period of time. Resources are increased. However, there is a problem that the equipment and resources to be increased become useless that is not normally used.

On the other hand, in the second method, an AP whose data has not been collected is randomly selected, and the AP
Data is collected from. However, for example, if the AP selection timing is in the second half of one cycle, but the length of time required to collect data from the AP is more than half of one cycle, the data will be within one cycle. There was a problem that the collection did not end.

  It is an object of the present invention to provide a technique that can complete data collection from a plurality of target devices within one period.

  One aspect of the present invention is a data collection device that collects data stored in a plurality of target devices by a plurality of processes that allow parallel access to the plurality of target devices. The data collection device includes an acquisition unit that assigns target devices to be accessed to the plurality of processes and acquires the data in order of long data collection time assumed for each target device.

  According to the present invention, information collection from a plurality of target devices can be completed within one cycle.

FIG. 1 shows an example of a data collection system. FIG. 2 is an explanatory diagram of an AP allocation method in data collection. FIG. 3 is an explanatory diagram of management of association data. FIG. 4 is a diagram illustrating an example in which the above-described first method is applied and a first case in which the problem occurs. FIG. 5 is an explanatory diagram of a second case in which a problem occurs in the first method. FIG. 6 is an explanatory diagram of an application example of the second method and its problem. FIG. 7 is an explanatory diagram of a data collection method according to the embodiment. FIG. 8 shows an example in which the data collection method of the embodiment is applied to the second case used for the description of the first method. FIG. 9 schematically illustrates a configuration example of the data collection system according to the embodiment. FIG. 10 shows a hardware configuration example of the collection management server. FIG. 11 shows a hardware configuration example of the collection server. FIG. 12 shows a hardware configuration example of the access point. FIG. 13 shows an example of the data structure of the collection target AP list table (list) shown in FIG. FIG. 14 shows an example in which an entry for a certain access point (AP # 2) is added to the collection target AP list table. FIG. 15 shows an example in which an entry related to AP # 2 is stored (registered) in the special designation table of the collection target AP list table. FIG. 16 is a flowchart illustrating an example of processing performed by the information acquisition unit (CPU) of each collection server. FIG. 17 is a flowchart illustrating an example of an assumed time determination process. FIG. 18 is a flowchart illustrating another example of the assumed time determination process. FIG. 19 is a diagram illustrating an example of the operational effect of the embodiment.

  Hereinafter, embodiments will be described with reference to the drawings. The configuration of the embodiment is an exemplification, and is not limited to the configuration of the embodiment.

In the embodiment, as an example, a wireless LAN on an Internet Protocol (IP) network
An information collection method of the access point (AP) of (Local Area Network) will be described.
In the embodiment, as an example, the collection server collects information using Simple Network Management Protocol (SNMP). The wireless LAN is an example of a wireless network, and the wireless network includes, for example, a wireless LAN that conforms to IEEE 802.11 series standards and WiFi standards. However, the wireless network may include a wireless network based on other standards. SNMP is an example of a protocol used for information collection, and a protocol other than SNMP can be applied.

  An access point (AP) is one of the wireless devices that enables wireless terminals to be connected to each other and to connect an unrelated terminal to a network. The wireless terminal includes, for example, a laptop personal computer (PC), a tablet terminal, a smartphone, a feature phone, and a personal digital assistant (PDA).

  FIG. 1 shows an example of a data collection system. The collection system includes one or more collection servers 10 and a collection management server 11. Each collection server 10 polls each of a plurality of access points (AP) 12 and collects data stored (accumulated) in the AP 12. As a method for collecting information from the AP 12, for example, there is a method in which the collection server 10 receives an SNMP trap notification transmitted from the AP 12. However, reception of the SNMP trap notification is not essential, and other collection methods can be applied.

  Each collection server 10 sends the collected data to the collection management server 11. The collection management server 11 includes a database (DB) 11 a that stores data collected by each collection server 10.

  The collection server 10 anticipates the amount of information collected from each AP (the time required for collection (also referred to as “collection time”)) at the peak time (when the maximum number of AP users increases or when the load becomes maximum). A number of APs are assigned as APs 12 for data collection.

  In other words, the peak collection time (collected information amount) is assumed, and the time length of one cycle (one cycle) for data collection is determined so that the data collection is completed with a margin. Then, the number of APs that each collection server collects data is set so that the collection time of each collection server is within a period of time. When the number of APs increases and the existing number of collection servers cannot collect data from all APs within one cycle, the number of collection servers is increased.

  When data is collected from a plurality of APs 12 by a plurality of collection servers 10, there are a first method and a second method for assigning APs 12 to each collection server 10. As shown in FIG. 2, in the first method, in consideration of the processing capability of each collection server 12, a number of APs 12 considered to be able to finish data collection within one cycle are fixedly assigned.

  In the second method, the collection management server 11 manages the collection status of data from the AP 12 by the collection server 10, randomly assigns the AP 12 that has not been collected to the collection management server 11, and the collection server 10 to which the AP 12 is assigned. Collects data.

  Here, association data (connection information for each terminal), which is an example of data collected from the AP 12 (data to be collected), will be described. The association data is data indicating the usage status of the AP 12. FIG. 3 is an explanatory diagram of management of association data.

As shown in FIG. 3, when the wireless terminal 13 moves into the wireless area 12a (referred to as “cell”) of the AP 12, connection processing between the wireless terminal 13 and the AP 12 is performed. A procedure in which the wireless terminal 13 is connected to the AP 12 and data communication is started between the two is called “association”. When the procedure is started, the Media Access Control (MAC) address of the wireless terminal 13 and
The terminal connection time indicating the elapsed time since the establishment of the association is stored in the AP 12 as association data. The association data (MAC address and terminal connection time) is stored as an entry in the table (AP collection information table) 14, for example.

  When the wireless terminal 13 moves out of the wireless area 12a, the connection between the wireless terminal 13 and the AP 12 is disconnected, and the corresponding entry is deleted from the table 14 after a predetermined time has elapsed. That is, an aging process is performed. An aging timer for the table 14 is provided for the aging process, and each time the aging timer expires, the entry of the wireless terminal 13 that does not exist in the wireless area 12a is deleted. The size of the table 14 (the number of entries that can be stored) has an upper limit. When the number of entries registered in the table 14 reaches the upper limit, the entries are overwritten in order from the oldest entry.

  When the collection server 10 collects association data, the collection server 10 acquires association data from the AP 12 before an entry is deleted from the table 14. For this reason, the collection server 10 completes polling for each AP 12 within the aging period of the AP 12, that is, before the aging timer expires.

  Here, considering that the old entry is overwritten when the table 14 becomes full, if the data collection cycle is shortened (the collection efficiency of the collection server 10 is improved), more entries can be collected at the time of congestion. There is sex.

  The amount of information in the association data depends on the traffic of people near the AP 12, that is, the number of wireless terminals 13 existing in the wireless area of the AP 12 and the usage status of the wireless terminals 13. That is, as the number of wireless terminals 13 that exist in the wireless area 12a of the AP 12 and communicate with the AP 12 increases, the number of entries of association data stored in the table 14 (that is, the amount of information collected by the collection server 10). Will increase.

  FIG. 4 is a diagram illustrating an example in which the above-described first method is applied and a first case in which the problem occurs. In FIG. 4, there is a collection server 10a as one of the collection servers 10, and APs (1) to (10) are fixedly assigned to the collection server 10a as APs 12 from which association data is collected.

  The collection server 10 in the first method executes processes for collecting association data from a plurality of APs 12 in parallel. Depending on the situation, one collection server 10 may collect association data from hundreds of thousands of APs in one cycle. In such a case, if a large amount of information transmitted from the AP 12 by User Datagram Protocol (UDP) is received, the buffer of the collection server 10 may overflow. For this reason, the amount of information collected by the collection server 10 at a time is limited by the number of processes, and within a period of time, each of a plurality of processes polls a predetermined number (for example, several) of APs to obtain association data. collect.

  In the example illustrated in FIG. 4, the collection server 10 a executes a predetermined number (5 in the example of FIG. 4) of processes 1 to 5. Process 1 collects association data from AP (1) and AP (6) within a period of time to collect association data. Process 2 collects association data from AP (2) and AP (7) within a period of time. Process 3 collects association data from AP (3) and AP (8) within a period of time. Process 4 collects association data from AP (4) and AP (9) within a period of time. Process 5 collects association data from AP (5) and AP (10) within a period of time.

  However, the first method may cause the following problems. In FIG. 4, it is assumed that the wireless terminal 13 is congested in the wireless area of the AP (9) due to an event or the like, and a large amount of association data is accumulated in the table 14 of the AP (9). In this case, the process 5 of the collection server 10 may not finish the collection of the association data from the AP (9) within a period of one cycle.

  As described above, since the association data is deleted from the table 14 by the aging process, association data that could not be collected within one cycle time cannot be collected in the next cycle. Further, as described above, it is preferable that the time length of one cycle is short. Therefore, in order to collect association data without omission, the processing capacity of the collection server 10a is improved or the collection server 10 for AP (9) is increased so that the collection can be completed within one specified period. Equipment and capacity will be enhanced.

  FIG. 5 is an explanatory diagram of a second case in which a problem occurs in the first method. In the second case shown in FIG. 5, an installation area of the AP 12 in the city center and an installation area of the AP 12 in the suburbs are considered. During the daytime hours on weekdays, people move from the suburbs to the city center. On the other hand, on weekday nights and on holidays, there is little movement of people from the suburbs to the city center, and the population in the suburbs is larger than the population in the city center.

  In the first method, the fixed AP 12 is assigned without considering the movement of the person described above. For this reason, the processing capacity and the number of the collection servers 10 are prepared so that association data can be collected with a margin even in the congestion peak of the wireless area of the AP 12 during the daytime on weekdays in the city center. In both the first case and the second case, the increase of the collection server 10 in accordance with the congestion peak maintains a wasteful resource that is not normally used.

  FIG. 6 is an explanatory diagram of an application example of the second method and its problem. In FIG. 6, a plurality of collection servers 10 (collection servers 10 b, 10 c, and 10 d in FIG. 6) collect association data for a plurality of APs 12 arranged in the city center and suburbs. Each of the collection servers 10b, 10c, and 10d performs collection by randomly selecting an AP 12 that has not yet collected association data in accordance with an instruction from the collection management server 11 (not shown in FIG. 6), for example.

  However, as in the process 5 of the collection server 10b in FIG. 6, a large amount of association data is accumulated in a certain AP 12 (assumed to be AP (x)), and the collection of association data does not end within a period of time. Cases can happen.

  As described above, in both the first method and the second method, there may occur a case where data collection does not end within one cycle time. For this reason, it is inefficient to reinforce the collection server that is normally wasted.

  In the embodiments described below, a data collection method and data that can reduce the scale of collection servers (the number of collection servers and the processing capability of each collection server) by completing data collection within one cycle. A collection device and a data collection system will be described.

The data collection device according to the embodiment collects data stored in a plurality of target devices by a plurality of processes capable of parallel access to the plurality of target devices (for example, at least two of the plurality of target devices). It is a data collection device. The data collection device includes an acquisition unit that assigns target devices to be accessed to the plurality of processes and acquires the data in order of long data collection time assumed for each target device. Parallel access means that each of a plurality of processes executes access to a target device assigned to itself in parallel.

  In the description of the embodiment, an example in which an example of the data collection device is the collection server 10A and an example of the target device is the AP 12 will be described. However, the data collection device may be a device other than the collection server. Further, the target device may be a device other than the AP as long as it stores data to be collected. Further, the data to be collected is, for example, association data stored in the AP 12, but may be data other than association data.

  FIG. 7 is an explanatory diagram of a data collection method according to the embodiment. In FIG. 7, for example, the upper case of FIG. 7 shows a first case using the first method described with reference to FIG. In the first method, only the AP 12 from which the processes 1 to 5 of the collection server 10 collect data is determined. The time required to collect data from each AP 12, that is, the data collection time is not considered. As a result, like the data collection time of AP (9), the end of the period is later than the end of one cycle, and data that cannot be collected is generated.

  On the other hand, in the embodiment, for each AP 12, a time related to data collection (acquisition) is assumed for each period. The collection server 10 </ b> A according to the embodiment assigns target devices to be accessed to a plurality of processes in descending order of data collection time in each assumed target device, and each process accesses each AP 12 to acquire data. Processes 1 to 5 are examples of “a plurality of processes capable of accessing the plurality of target devices in parallel”.

  For example, the lower part of FIG. 7 shows the allocation result for each process of each AP when the data collection method according to the embodiment is applied to the first case. As shown in FIG. 7, the data collection time for each of the APs (1) to (10) has a length (time length) equivalent to the data collection time shown in the first method (upper part of FIG. 7). Assume that there is.

  In this case, the collection server 10A sorts the APs in the order in which the data collection time is long, and distributes each AP to the processes 1 to 5. As a result, among APs (1) to (10), AP (9) having the longest data collection time is assigned to the first process 1. Therefore, data collection for AP (9) is started from the beginning of one cycle. As a result, data collection for the AP (9) can be completed (completed) before the end of one cycle.

  The remaining APs (1) to (8) and (10) are also assigned to the processes in the order of the longest assumed data collection time. In the example shown in FIG. 7, the data collection time in each AP (1), (3), (4), (7), (8) is the same, and each AP (2), (5), (6) , (10) has the same data collection time, and the latter time length is half the translation of the former time length.

  When the data collection time is the same length, for example, the AP 12 is assigned to the process in the ascending or descending order of the numbers previously assigned to the AP 12. In the example of FIG. 7, APs (1), (3), (4), and (7) are assigned to the processes 2 to 5 in ascending order from the smallest number. When the next AP (8) is assigned to process 1 (after AP (9)), the end of data collection time is later than the end of one cycle, so it is assigned to process 2 (after AP (1)). It has been. The remaining APs (2), (5), (6), and (10) are assigned to processes 2 to 5 in the same manner.

According to the data collection method according to the embodiment, it is possible to finish data collection of all the APs (1) to (10) that are data collection target devices (target devices) within one cycle of data collection. As a result, data can be collected without omission even if the collection server is not increased.

  In addition, according to the data collection method according to the embodiment, the time when data collection from all of the plurality of target devices (the plurality of APs 12 (AP (1) to (10)) is completed is compared with the first method. Therefore, depending on the assumed data collection time, the time length of one cycle for data collection can be shortened, in other words, more data (association). Data entry).

  In the above description, the collection server 10A has been described as performing the assignment of the AP 12 to each process. However, the assignment of the AP 12 to each process of each collection server 10A may be performed by the collection management server 11 (not shown in FIG. 7). In this case, the allocation result in the collection management server 11 is notified to each collection server 10A, and each collection server 10A collects data according to the allocation result.

  FIG. 8 shows an example in which the data collection method of the embodiment is applied to the second case used for the description of the first method. As illustrated in FIG. 8, the collection server 10A selects APs in the descending order of the assumed data collection time. As a result, as shown in FIG. 8, data collection from the plurality of APs 12 is completed within one cycle time, whether it is a daytime daytime on weekdays, a nighttime day on weekdays, or a holiday. APs are assigned as described above.

  Furthermore, according to the data collection method according to the embodiment, as shown in FIG. 8, with respect to the collection server 10 </ b> A, the load of data collection performed during weekday daytime and the weekday nighttime period and holidays are performed. The data collection load can be made comparable. Thus, the number of collection servers 10A required for data collection can be reduced by efficiently assigning the data collection processing to the plurality of collection servers 10A. In other words, it is possible to reduce the scale of the collection server and to reduce the cost for maintenance.

<Configuration example of data collection system>
FIG. 9 schematically illustrates a configuration example of the data collection system according to the embodiment. In FIG. 9, the data collection system includes a collection management server 11A and a plurality of (m: m is a positive integer) collection servers 10A (# 1 to #m) connected to the collection management server 11A via a network. Including. The data collection system includes a plurality of (n: n is a positive integer) APs 12 (# 1 to #n) connected to each collection server 10A via a network.

  The AP 12 includes an access point main function (main function) 121, an information management unit 122, a collected information table 123, and an information transmission unit 1245. The main function 121 forms a wireless area (cell) by radiating radio waves and transmits notification information to the wireless area. The wireless terminal 13 that exists in the wireless area and receives the broadcast information performs a procedure for establishing an association between the AP 12 and the wireless terminal 13 using the broadcast information in order to communicate with the AP 12, and performs the association. Establish.

  The information management unit 122 accumulates and manages information obtained by access from the wireless terminal 13 in the collected information table 123 each time. For example, the information management unit 122 stores association data generated by establishing an association (connection with the wireless terminal 13) with the wireless terminal 13 in the collection information table 123.

The collection information table 123 is the same as the table 14 illustrated in FIG. 3, and includes, for example, the MAC address of the wireless terminal 13 and the terminal connection time of the wireless terminal 13 for each wireless terminal 13 that has established an association. Store the entry. The information management unit 122 writes, updates, and deletes association data with respect to the collected information table 123. The information management unit 122 performs the aging process described above. Also, when the number of registered entries in the collection information table 123 reaches the upper limit, overwriting is performed in order from the oldest entry.

  In response to polling from the collection server 10A, the information transfer unit 124 reads association data stored (accumulated) in the collection information table 123 and transmits the association data to the collection server 10A.

  The collection server 10A includes an information acquisition unit 101 and an information storage unit 102. The information acquisition unit 101, in a predetermined cycle for data collection (also referred to as “polling cycle”), for a plurality of APs 12 (# 1 to #n), in order from the longest assumed data collection time, AP 12 is assigned to a plurality of processes 1 to 5, polling (access) is performed, and association data is collected. The collection server 10A accesses the plurality of APs 12 based on the collection target AP list table 112 managed by the collection management server 11A. The information acquisition unit 101 is an example of an “acquisition unit”.

  The information storage unit 102 transmits the association data acquired by the information acquisition unit 101 to the collection management server 11A. The association data is stored in the information accumulation table 113 included in the collection management server 11A.

  The collection management server 11A includes a collection time management unit 111, a collection target AP list table 112 (hereinafter referred to as a list 112), and an information accumulation table 113. The collection time management unit 111 presets the data collection time assumed for each AP 12 in the list 112 in advance. The list 112 is an example of a “management unit”, and the information accumulation table 113 is an example of an “accumulation unit”.

  The list 112 stores, for example, an identifier of the AP 12, a data collection time assumed for the AP 12 (assumed time), and a status indicating whether data has been collected or not collected. The estimated time is set, for example, by an administrator of the data collection system through an experiment or statistical processing of past actual values. The information acquisition unit 101 of the collection server 10A accesses the list 112 of the collection management server 11A and determines the access order (polling order) of the AP 12. In the information accumulation table 113, data collected by each collection server 10A is received from each collection server 10A and accumulated. The AP 12 is assigned to the plurality of processes 1 to 5 by, for example, the process 1 → 2 → 3 → 4 → 5 → 1. . . Of round robin.

  FIG. 10 shows a hardware configuration example of the collection management server 11A. As the collection management server 11A, for example, a general-purpose computer such as a personal computer (PC) or a workstation (WS) can be applied. Alternatively, a dedicated computer such as a server machine can be applied as the collection management server 11A.

As shown in FIG. 10, the collection management server 11A includes, for example, a central processing unit (CPU) 21, a memory 22, a storage medium 23, and a line interface (line I / F) connected to each other via a bus. ) 24. The memory 22 is used as a main storage device, and is used as a program development area, a CPU 21 work area, and a data buffer (storage) area. The memory 22 is formed by, for example, a random access memory (RAM) or a combination of a RAM and a read only memory (ROM).

The storage medium 23 is used as an auxiliary storage device. The auxiliary storage device is formed by a non-volatile storage medium such as a hard disk drive (HDD), a solid state drive (SSD), a flash memory, or an electrically erasable programmable read-only memory (EEPROM). The storage medium 23 is used as a storage area for data and programs. The list 112 and the information accumulation table 113 described above are stored in at least one of the memory 22 and the storage medium 23.

  The line interface 24 is connected to the network and manages communication between the collection management server 11A and each collection server 10A. As the line interface 24, for example, a communication interface card called a local area network (LAN) card or a network interface card (NIC) can be applied.

  The CPU 21 is an example of a processor, and loads a program stored in the storage medium 23 into the memory 22 and executes it (the program (firmware) stored in the ROM included in the memory 22 may be executed).

  Accordingly, the CPU 21 operates as the collection time management unit 111 illustrated in FIG. In the operation as the collection time management unit 111, the CPU 21 communicates with the collection server 10 </ b> A using the line interface 24.

  FIG. 11 shows a hardware configuration example of the collection server 10A. As the collection server 10A, for example, the above-described PC, WS, server machine, or the like can be applied. As shown in FIG. 11, the collection server 10A includes a CPU 31, a memory 32, a storage medium 33, and a line interface (line I / F) 34 that are connected to each other via a bus.

  The functions of the memory 32, storage medium 33, and line interface 34 are the same as those of the memory 22, storage medium 33, and line interface 24. Similar to those described for the memory 22, the storage medium 23, and the line interface 24, the memory 32, the storage medium 33, and the line interface 34 can be applied.

  The CPU 31 is an example of a processor, and loads a program stored in the storage medium 33 to the memory 32 and executes it (the program (firmware) stored in the ROM included in the memory 32 may be executed).

  Thus, the CPU 31 operates as the information acquisition unit 101 and the information storage unit 102 illustrated in FIG. In the operation as each of the information acquisition unit 101 and the information storage unit 102, the CPU 31 communicates with the AP 12 or the collection management server 11A using the line interface 34. In addition, the CPU 31 generates a plurality of processes (processes 1 to 5) and collects data by the processes 1 to 5. The CPU 31 assigns the AP 12 to the processes 1 to 5 and causes each process to collect data from the AP 12.

  FIG. 12 shows a hardware configuration example of the access point (AP) 12. As shown in FIG. 12, the AP 12 includes a CPU 41, a memory 42, a storage medium 43, a line interface (line I / F) 44, and a wireless interface 45 connected to each other via a bus.

  The functions of the memory 42, the storage medium 43, and the line interface 44 are the same as those of the memory 22, the storage medium 33, and the line interface 24. The same as those described for the memory 22, the storage medium 23, and the line interface 24 can be applied as the memory 42, the storage medium 43, and the line interface 44. The collected information table 123 is stored in at least one of the memory 42 and the storage medium 43.

The wireless interface 45 manages wireless communication with the wireless terminal 13. Wireless interface 4
5 includes, for example, a baseband circuit that performs a conversion process between data and a baseband signal, a radio frequency (RF) circuit that performs a conversion process between the baseband signal and the radio signal, and an antenna.

  The CPU 41 is an example of a processor, and loads a program stored in the storage medium 43 or the ROM of the memory 42 into the memory 42 and executes it. Accordingly, the CPU 41 can operate as the main function 121, the information management unit 122, and the information transmission unit 124 illustrated in FIG.

  The CPU 41 operating as the main function 121 controls, for example, the operation of the wireless interface 45, radiates radio waves from the wireless interface 45 to form a wireless area, transmits notification information, and receives a connection request from the wireless terminal 13 ( Wait for association establishment request. When a connection request is received from the wireless terminal 13 that has received the broadcast information, an association establishment procedure is established with the wireless terminal 13. After the establishment, wireless data communication is performed with the wireless terminal 13.

  The CPU 41 operating as the information management unit 122 stores the MAC address and terminal connection time of the wireless terminal 13 with which the association has been established in the collected information table 123. The CPU 41 operating as the information transmission unit 124 communicates with the collection server 10 </ b> A using the line interface 44.

Note that at least some of the processes executed by the CPU 21, CPU 31, and CPU 41 may be executed using hardware. Hardware is Field Programmable
A programmable logic device (PLD) such as a gate array (FPGA) or an integrated circuit (IC, LSI, Application Specific Integrated Circuit (ASIC)) is included.

  FIG. 13 shows an example of the data structure of the collection target AP list table 112 (list 112). In FIG. 13, a list 112 includes a collection target AP list table (list) 112a for weekdays, a collection target AP list table (list) 112b for holidays, and a specially specified collection target AP list table (list) 112c. Contains. The list 112 can be held in the DB 11a in the collection management server 11A.

  The list 112a and the list 112b have the same data structure. That is, each list includes one or more records including “collection period”, “AP number”, “assumed collection time”, “actual collection time”, and “implementation status” on the current day (weekdays or holidays). (Entry) is stored.

  As the “collection cycle”, the start time of one cycle of data collection is stored. The time length of one cycle is determined in advance. For example, the time length of one cycle is set to 5 minutes, but the time length can be set as appropriate. “AP number” is unique identification information of the AP 12.

  The “assumed collection time” indicates the time assumed to be required for collecting data from the AP 12 specified by the AP number. As the assumed collection time, for example, a value determined by the administrator or maintenance person of the data collection system is set. The “actual collection time” stores the time required for actual data collection. “Implementation status” indicates the implementation status of data collection in one cycle specified by the collection cycle. In the embodiment, as the implementation status, “not collected” indicating that data collection is not being performed, “collecting” indicating that data collection is being performed, and “collected” indicating that data collection has been completed. Is set.

  In each of the list 112a and the list 112b, records (entries) corresponding to a plurality of collection cycles performed on the same day are stored. On the other hand, the start time of the event is stored in the list 112c as “collection cycle date and time”. However, the list 112c is the same as the lists 112a and 112b in that one record (entry) includes “AP number”, “assumed collection time”, “actual collection time”, and “implementation status”. .

  FIG. 14 shows an example of adding an entry for AP # 2 to the list 112a and the list 112b. In this case, as shown in FIG. 14, the collection time management unit 111 sets a collection period, an AP number, an assumed collection time, an actual collection time (initial values, for example, the same values as the assumed collection time) in the list 112a and the list 112b. ), And setting the implementation status.

  FIG. 15 shows an example in which an entry related to AP # 2 is stored (registered) in the list 112c. In the list 112c, an assumed collection time in which an area where an event occurs and a time (an event occurring at a certain time in a certain area) is considered in advance is set.

  As shown in FIG. 15, for example, in an area where AP # 2 exists, the user of the wireless terminal 13 may change between 23:55 on December 31, 2014 and 0:00 on January 1, 2015. If it increases, an entry including the estimated collection time is set in the list 112c.

  FIG. 16 is a flowchart illustrating a processing example performed by the information acquisition unit 101 (CPU 31) of each collection server 10A. The process of FIG. 16 is periodically activated in accordance with the data collection period.

  In 01, the information acquisition unit 101 searches the collection target AP list table 112 (list 112) of the collection management server 11A for an AP 12 having a long “expected collection time”. In 01, when retrieving the list 112, the information acquisition unit 101 uses a list to be searched depending on whether the collection date (the current day) is a weekday or a holiday from the list 112a and the list 112b. However, the search is performed including the list 112c.

  As a search condition, a search is performed in which the time of the collection cycle is the cycle of the current time, the implementation status is “not collected”, and the estimated collection time is the longest. At this time, the presence or absence of an entry that matches the current time condition is searched from the special designation (list 112c).

  In 02, the information acquisition unit 101 determines whether a search result (an entry that matches the search condition) has been obtained. When no search result is obtained (02, NO), that is, when there is no AP 12 that collects data in the collection cycle, the process of FIG. 16 ends. If there is a search result (02, YES), the process proceeds to 03.

  In 03, the information acquisition unit 101 sets “collecting” to the entry retrieved from the list 112. In 04, the information acquisition unit 101 assigns the AP 12 having the AP number included in the searched entry to the process. The information acquisition unit 101 manages a plurality of processes 1 to 5 and assigns the AP 12 specified by the entry searched in 03 to the process to which the AP 12 is assigned next. The process to which the AP 12 is assigned accesses the assigned AP 12 and acquires data stored in the collection information table 123.

  In 05, when data collection is distorted, the information acquisition unit 101 sets “collected” as the implementation status in the searched entry, and sets the actual data collection time to “actual collection time” in the entry.

In 06, the information acquisition unit 101 sets the collection status in the entry searched for 01 in the list 112 to “uncollected”. In this case, the collection status is set to “not collected” in advance for data collection in the next same time zone. In 07, the information acquisition unit 101 performs an assumed time determination process (subroutine) and returns the process to 01.

  FIG. 17 is a flowchart illustrating an example of an assumed time determination process. In the process of 07a in FIG. 17, the information acquisition unit 101 sets the “real collection time” of the current cycle to the “assumed collection time” of the next cycle. For example, as shown in FIG. 17, the cycle starts every 5 minutes, and the cycle entry starts at 00:05 (referred to as entry 1), and the cycle entry starts at 00:10 (entry 2 and )).

  In entry 1, when the assumed collection time is set to 10 seconds (10 s), but the actual collection time is 5 seconds (5 s), the information acquisition unit 101 sets the actual collection time of the current cycle. , Set to the assumed collection time of entry 2 in the next cycle. The process of 07a is an example of “use the actual data collection time in a certain cycle as the data collection time assumed for the next cycle”.

  When it is considered that there is no significant difference in the number of wireless terminals 13 existing in the wireless area of the AP 12 between the current period and the next period, the process 07a is executed. As a result, the assumed collection time can be made closer to the actual, and the AP selection order can be accurately performed.

  FIG. 18 is a flowchart illustrating another example of the assumed time determination process. In the process of 07b in FIG. 18, the information acquisition unit 101 sets the average of “actual collection time” from Monday to Friday in the next cycle as “assumed collection time” in the next cycle.

  For example, as shown in FIG. 18, the actual collection time from Monday to Friday is stored in the entry of the list 112. In the example of FIG. 18, the actual collection time is managed for each day of the week. In the process 07b, the information acquisition unit 101 calculates the average value of the actual collection time, and sets the average value to the “assumed collection time” of the next period. As a result, the influence when the actual collection time of this cycle is a singular value can be reduced. In the processing of 07b, “the data collection time assumed based on the result of statistical processing of the data collection times of the plurality of target devices in the past two or more cycles is used as the data collection time assumed for the next one cycle. Is an example.

  Alternatively, as another assumed collection time setting process, it is conceivable to set the actual collection time obtained in the same period as the next cycle as the assumed collection time of the next cycle, obtained last week. The collection time assumed by using a method other than these may be set as the assumed collection time.

  When the assumed collection time is set in 07a or 07b, entry information is transmitted from the collection server 10A to the collection management server 11A. The collection time management unit 111 (CPU 21) stores the entry in the list 112. When the processing of 07 is completed, the processing returns to 01 and the next AP 12 is determined.

<Effects of Embodiment>
The information acquisition unit 101 repeats the processing illustrated in FIG. 16 and performs data collection for all APs 12 registered in the current cycle entry in the list 112 in the order of long assumed collection time corresponding to the AP 12. As described above, the information acquisition unit 101 is an example of “an acquisition unit that allocates target devices to be accessed to the plurality of processes in the order of long data collection time assumed for each target device and acquires the data”. It is.

According to the embodiment, by performing the above-described data collection, it is possible to finish collecting data from all of the plurality of APs 12 that collect data within one cycle. That is, information collection from a plurality of target devices can be completed within one cycle. For this reason, it is possible to avoid the generation of useless resources and facilities by avoiding the increase of the collection servers as described in the first method and the second method.

  For example, in the first method, as illustrated in FIG. 5, the collection server 10 is provided in consideration of the peak information amount in each of the suburbs and the city center. On the other hand, according to the data collection method of the embodiment, by installing the collection server 10A that can process the information amount in the peak time zone in the suburbs and the city center, the processing efficiency is improved and the number of installations is increased. Can be suppressed (FIG. 8).

  Furthermore, an entry including an assumed collection time as a data collection time assumed based on an event occurring at a certain time in a certain region is registered (stored) in the list 112c and used for searching for the AP 12 that is a data collection target. The As a result, data collection can be performed in accordance with an increase in the number of wireless terminals 13 due to irregular events, and omissions in acquisition can be suppressed.

  For example, as shown in FIG. 19, a relatively long assumed collection time is provided for the AP 12 (assumed to be AP 12x) in the event venue in the event occurrence cycle, and the AP 12x is accumulated in one cycle. Allow time to collect large amounts of data.

  In the embodiment, the collection management server 11A centrally manages the list 112, but each collection server 10A may individually have a list for the data collection target AP 12. Further, the assignment of the AP 12 to a plurality of processes in each collection server 10 may be performed by, for example, a device other than the data collection device, for example, the collection management server 11A. At this time, the collection management server 11A (an example of a device other than the data collection device) may perform allocation to a part of the plurality of collection servers 10, or may perform allocation to all of the plurality of collection servers 10. good. In this case, in the collection server 10, for example, the CPU 31 refers to the assignment result in the collection management server 11A, assigns the AP 12 to each process of the collection server 10, and each process acquires data from the AP 12 assigned to itself. . The configurations of the embodiments described above can be combined as appropriate.

10A ... collection server 11A ... collection management server 12 ... access point 13 ... wireless terminals 21, 31 ... CPU
22, 32 ... Memory 23, 33 ... Storage medium 101 ... Information acquisition unit 111 ... Collection time management unit 112 ... Collection target AP list table

Claims (6)

In a data collection device for collecting data stored in a plurality of target devices by a plurality of processes capable of parallel access to the plurality of target devices,
An acquisition unit that acquires the data by assigning target devices to be accessed to the plurality of processes in order of long data collection time assumed for each target device;
Data collection device. The data acquisition apparatus according to claim 1, wherein the acquisition unit uses an actual data collection time in a certain cycle as a data collection time assumed for the next cycle. The acquisition unit uses a data collection time assumed based on a result of statistical processing of data collection times of the plurality of target devices in two or more past cycles as a data collection time assumed for the next one cycle. Item 4. The data collection device according to Item 1. The data acquisition device according to claim 1, wherein the acquisition unit uses a data collection time assumed based on an event that occurs at a certain time in a certain region. A plurality of data collection devices for collecting data stored in a plurality of target devices by a plurality of processes capable of parallel access to the plurality of target devices;
A management device for the plurality of data collection devices,
Each of the plurality of data collection devices includes:
An acquisition unit that acquires the data by allocating target devices to be accessed to the plurality of processes in order of long data collection time assumed for each target device;
The management device includes a management unit that manages the assumed data collection time, and a storage unit that stores data acquired by the plurality of data collection devices. A data collection device that collects data stored in a plurality of target devices by a plurality of processes capable of parallel access to at least two of the plurality of target devices, in ascending order of data collection time assumed for each target device. Assigning a target device to access to the plurality of processes to obtain the data;
Data collection method.