JP3978627B2 - Data collection method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a data collection method for collecting data of a plurality of terminal units by a management machine via a plurality of intermediate units.
[0002]
[Prior art]
A wiring block diagram representing a conventional data management system in this regard is shown in FIG. This data management system is a system for summing up data of pachinko machines, and each pachinko machine (P machines) is equipped with a terminal unit (DU) that sums up the data. The data of the terminal unit (DU) is collected by the intermediate unit (SU). The intermediate unit (SU) is a unit that efficiently collects data of a plurality of terminal units (DU), and data of the intermediate unit (SU) is collected by a management machine. The management machine is a unit that collectively manages several intermediate units (SU), and a predetermined number is installed according to the scale of the store. Here, each intermediate unit (SU) is independent and can collect data of terminal units (DU) within the assigned range.
[0003]
[Problems to be solved by the invention]
However, according to the data management system, since each intermediate unit (SU) is independent, if one intermediate unit (SU) fails, the data of the terminal unit (DU) related to the intermediate unit (SU) is lost. There is a problem that it cannot be collected at all.
[0004]
Accordingly, the object of the present invention is to improve fault tolerance by enabling another intermediate unit to back up the failure range even when one intermediate unit fails. It is what.
In addition to the object of the invention described in claim 1, the object of the invention described in claim 2 is to prevent the data collection time from increasing even if each intermediate unit is backed up. It is.
Further, an object of the present invention is to shorten the data collection time and improve the fault tolerance of the system.
[0005]
[Means for Solving the Problems]
The above-described problem is solved by a data collection method having the following features.
That is, the data collection method according to claim 1 is a data collection method for collecting data of a plurality of terminal units by a management machine via a plurality of intermediate units, wherein the terminal units and the intermediate units are connected by a communication line. The data of each terminal unit is collected by at least two intermediate units with different priorities , and the intermediate unit is lower than the terminal units that collect high priority data with respect to a plurality of connected terminal units. The data is divided into terminal units that collect priority data, the data of each terminal unit is collected at each divided priority, and the data of each terminal unit collected at a different priority is transmitted to the management machine.
[0006]
According to the present invention, since data transmitted from each terminal unit is collected by at least two intermediate units, even if one intermediate unit fails, the data of terminal units within that range is different. It can be collected by the management machine via the intermediate unit. For this reason, the fault tolerance of the system is improved.
In addition, since the data of each terminal unit is collected by at least two intermediate units with different priorities, if the backup data is collected with a low priority, the data collection time will be greatly increased by the backup function. There is no.
[0007]
Further, the data collection method according to claim 2 is the data collection method according to claim 1,
Usually, the data collected by the high priority intermediate unit is collected by the management machine, and when the high priority intermediate unit fails, the data collected by the low priority intermediate unit is collected by the management machine.
[0008]
According to the present invention, since a low priority intermediate unit backs up only when a high priority intermediate unit fails, the data collection time is not increased by the backup function under normal conditions.
[0009]
The data collection method according to claim 3 is the data collection method according to claim 1 or 2,
The intermediate unit and the management device are connected by a communication line different from the communication line connecting the terminal unit and the intermediate unit.
[0010]
According to the present invention, since data can be transmitted between the intermediate unit and the management unit while data is being transmitted between the terminal unit and the intermediate unit, data collection time is shortened. The In addition, the fault tolerance of the system is improved.
[0011]
A data collection method according to claim 4 is the data collection method according to any one of claims 1 to 3, wherein a plurality of terminal units are divided into n collection ranges, and a plurality of intermediate units are divided. Data of terminal units within two collection ranges are collected in pairs of two units and in an intermediate unit of one pair.
One intermediate unit in the two intermediate units constituting the pair collects data at a high priority frequency from the terminal units in one collection range in the two collection ranges, and the high priority from the terminal unit in the other collection range The data is collected at a low priority frequency that is lower than the frequency, and the other intermediate unit in the two intermediate units that constitute the pair has a high priority frequency from the terminal unit in the other collection range in the two collection ranges. Data is collected, and data is collected from terminal units in one collection range at a low priority frequency that is lower than the high priority frequency.
The management device collects data of the terminal units in the collection range collected from each intermediate unit at the high priority frequency, and the faulty intermediate unit collects data from the terminal units in the collection range to be collected at the high priority frequency. The data is backed up with data from the terminal unit in the collection range collected at a low priority frequency from the intermediate unit that forms a pair with the failed intermediate unit.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a data collection method according to an embodiment of the present invention will be described with reference to FIG. The data collection method according to the present embodiment relates to a method of totalizing pachinko machine data, and FIG. 1 shows a main part wiring block diagram of the pachinko machine data management system.
Each pachinko machine (P machine) is equipped with a terminal unit (DU) that aggregates the data of those pachinko machines (P machine), and the data of those terminal units (No. 1DU to No. 500DU) The data is collected by No. 1 to No. 10 intermediate units (No. 1 SU to No. 10 SU) via the collection LAN.
[0013]
As shown in FIG. 1, the No. 1 intermediate unit (No. 1 SU) stores data of 50 terminal units (No. 1 DU to No. 50 DU) belonging to the first collection range (charge range). It is possible to collect data of 50 terminal units (No. 51 DU to No. 100 DU) belonging to the second collection range (backup range) with a low priority order. The No. 2 to No. 10 intermediate units (No. 2 SU to No. 10 SU) also function in the same manner as the No. 1 intermediate unit (No. 1 SU).
[0014]
Here, the No. 1 intermediate unit (No. 1 SU) and No. 2 intermediate unit (No. 2 SU) are used in pairs, and the first collection range (area in charge) of the No. 2 intermediate unit (No. 2 SU) The terminal units (No. 51DU to No. 100DU) belonging to No. 1 belong to the second collection range (backup range) of the No. 1 intermediate unit (No. 1 SU) at the same time. Conversely, the terminal units (No. 1DU to No. 50DU) belonging to the first collection range (charge range) of the No. 1 intermediate unit (No. 1 SU) are the No. 2 intermediate unit (No. 2 SU). Belongs to 2 collection range (backup range).
[0015]
Similarly, No. 3 intermediate unit (No. 3 SU) and No. 4 intermediate unit (No. 4 SU), No. 5 intermediate unit (No. 5 SU) and No. 6 intermediate unit (No. 6 SU), No. 7 The intermediate unit (No. 7 SU), No. 8 intermediate unit (No. 8 SU), No. 9 intermediate unit (No. 9 SU) and No. 10 intermediate unit (No. 10 SU) are used in pairs.
[0016]
Data collected by No. 1 to No. 10 intermediate units (No. 1 SU to No. 10 SU) is collected by the management machine via the management LAN. The management machine is a unit that collectively manages 10 intermediate units (SU), and a predetermined number is installed according to the scale of the store.
[0017]
Next, communication in the case where the intermediate units (SU) are provided independently of each other as in the prior art and in the case where two intermediate units (SU) can be paired and backed up as in the present embodiment Compare the processing time (collection time) and the load on the management machine.
Here, since the terminal unit (DU) and the intermediate unit (SU) are connected by the collection LAN, data of a single terminal unit (DU) can be communicated at a time. For this reason, assuming that it takes 0.1 second to communicate the data of one terminal unit (DU) to the intermediate unit (SU) and update the data of the intermediate unit (SU), the intermediate unit (SU) The data update time of the terminal unit (DU) at is a value obtained by multiplying the number of terminal units (DU) handled by the intermediate unit (SU) by 0.1 seconds.
[0018]
Similarly, since the intermediate unit (SU) and the management device are connected by a management LAN, data of a single intermediate unit (SU) can be communicated at a time. For this reason, assuming that it takes 0.5 seconds to communicate the data of one intermediate unit (SU) to the management machine and update the data of the management machine, the data update of the intermediate unit (SU) in the management machine The time is a value obtained by multiplying the number of intermediate units (SUs) in charge of the management machine by 0.5 seconds. Therefore, the time (collection time) for updating the data of all the pachinko machines managed by the management machine is [data update time of the terminal unit (DU) in the intermediate unit (SU)] and [intermediate unit (SU in the management machine). ) Data update time].
[0019]
In accordance with the above-described concept, the data update time in the management machine when the intermediate units (SU) are provided independently of each other as in the conventional case (see FIG. 2) is calculated. It is assumed that the intermediate unit (SU) manages 50 terminal units (DU), and the management machine manages 10 intermediate units (SU).
[Data update time of terminal unit (DU) in intermediate unit (SU)] = [Number of terminal units (DU)] × [0.1 seconds] = 50 units × 0.1 seconds = 5 seconds [intermediate in the management unit Unit (SU) data update time] = [number of intermediate units (SU)] × [0.5 seconds] = 10 units × 0.5 seconds = 5 seconds Therefore, [collection time] = 5 seconds + 5 seconds = 10 Second.
[0020]
Next, the update time of data in the management machine when the two intermediate units (SU) are paired and can be backed up with each other as in the present embodiment is calculated (see FIG. 1). Here, the data of the terminal unit (DU) belonging to the second collection range is collected at a frequency of 1/10 of the data of the terminal unit (DU) belonging to the first collection range. And For this reason, the number of terminal units (DUs) handled by the intermediate unit (SU) at a time is 50 in the first collection range and 50 × 1/10 in the second collection range, for a total of 55 units.
[0021]
Therefore, [data update time of terminal unit (DU) in intermediate unit (SU)] = [number of terminal units (DU)] × [0.1 seconds] = 55 units × 0.1 seconds = 5.5 seconds [ Data update time of intermediate unit (SU) in management machine] = [number of intermediate units (SU)] × [0.5 seconds] = 10 units × 0.5 seconds = 5 seconds Therefore, [collection time] = 5. 5 seconds + 5 seconds = 10.5 seconds. That is, the collection time of the management machine is about 5% longer than before.
[0022]
Next, the load on the management machine is calculated. Generally, the load on the high-speed LAN communication line and the management machine is more influenced by the number of packets required for communication than the total amount of messages to be communicated. For this reason, it is assumed that the load on the management machine or the like is grasped from the total number of packets.
Here, assuming that the data that can be communicated to the intermediate unit (SU) by the terminal unit (DU) is 100 bytes, the minimum packet length is 100 bytes, and the maximum packet length is 1000 bytes, the data of each terminal unit (DU) Can be sent in one packet. For this reason, the number of packets between the terminal unit (DU) and the intermediate unit (SU) is obtained by multiplying the number of terminal units (DU) handled by one intermediate unit (SU) by the number of intermediate units (SU). Is equal to
[0023]
Further, the value obtained by dividing the amount of data sent by one intermediate unit (SU) to the management machine by the maximum packet length (1000 bytes) is the number of packets per intermediate unit (SU). For this reason, the total number of packets between the intermediate unit (SU) and the management unit is equal to a value obtained by multiplying the number of packets per intermediate unit (SU) by the number of packets. When there are a plurality of management machines, the value is further multiplied by the number of management machines.
[0024]
In accordance with the above concept, the total number of packets when the intermediate units (SU) are provided independently of each other as in the conventional case is calculated (see FIG. 2).
First, the number of packets between the terminal unit (DU) and the intermediate unit (SU) is calculated.
Since the number of terminal units (DU) handled by one intermediate unit (SU) is 50 and the number of intermediate units (SU) is 10,
[Number of packets] = 50 × 10 = 500.
[0025]
Next, the number of packets between the intermediate unit (SU) and the management unit is calculated.
Since the amount of data sent from one intermediate unit (SU) to the management device is the number of terminal units (DUs) × 100 bytes, 50 units × 100 bytes = 5000 bytes. Therefore, the required number of packets is 5000 bytes ÷ maximum packet length (1000 bytes) = 5.
[0026]
Furthermore, since there are 10 intermediate units (SU), the total number of packets between the intermediate unit (SU) and the management unit is 10 × 5 = 50. When there are a plurality of management machines, the data of the intermediate unit (SU) is sent to each management machine, so the number of packets is 50 times the number of management machines.
Therefore, the total number of packets when the intermediate units (SU) are provided independently from each other as in the conventional case is 500 + 50 × the number of management machines.
[0027]
Next, the total number of packets when the two intermediate units (SU) are paired and can be backed up with each other as in the present embodiment is calculated (see FIG. 1).
First, the number of packets between the terminal unit (DU) and the intermediate unit (SU) is calculated.
The number of terminal units (DU) in charge of one intermediate unit (SU) is 50 in the first collection range and 50 in the second collection range, but in the second collection range, 1 / of the first collection range. Since data is collected at a frequency of 10, the number of terminal units (DUs) handled by one intermediate unit (SU) is considered to be 50 × 1.1 = 55. In addition, since the number of intermediate units (SU) is 10,
[Number of packets] = 55 × 10 = 550.
[0028]
Next, the number of packets between the intermediate unit (SU) and the management unit is calculated.
Usually, each intermediate unit (SU) only needs to send data collected in the first collection range between the intermediate unit (SU) and the management unit, so one intermediate unit (SU) sends it to the management unit. The amount of data is 50 units × 100 bytes = 5000 bytes. Therefore, the required number of packets is 5000 bytes ÷ maximum packet length (1000 bytes) = 5.
[0029]
Furthermore, since there are 10 intermediate units (SU), the total number of packets between the intermediate unit (SU) and the management unit is 10 × 5 = 50. When there are a plurality of management machines, the data of the intermediate unit (SU) is sent to each management machine, so the number of packets is 50 times the number of management machines.
Therefore, the total number of packets when two intermediate units (SU) are paired and can be backed up with each other as in this embodiment is 550 + 50 × the number of management machines.
[0030]
For this reason, the number of packets is increased by about 7% to 9% compared to the conventional case where the intermediate units (SU) are provided independently of each other. As the number of management machines increases, the rate of increase in the number of packets decreases.
As described above, the duplication for the failure of the intermediate unit (SU) can be realized by increasing the collection time by about 5% and increasing the number of packets by about 8%.
[0031]
According to this embodiment, since data transmitted from each terminal unit (DU) is collected by two or more intermediate units (SU), even if one intermediate unit (SU) fails, it is in charge The data of the terminal units within the range can be collected by the management machine via another intermediate unit (SU). For this reason, the fault tolerance of the system is improved. The failure of the intermediate unit (SU) can be known, for example, by monitoring the data transmission status of the intermediate unit (SU) within a predetermined time by the management device.
[0032]
In addition, in order to collect terminal unit (DU) data in the range to be backed up (second collection range) with a low priority, even if a backup function is provided, the data collection time will increase significantly compared to the conventional case. There is no (about 5% increase).
Further, since the collection LAN and the management LAN are separated, while data is being transmitted between the terminal unit (DU) and the intermediate unit (SU), between the intermediate unit (SU) and the management machine. Can transmit data, and the data collection time is shortened. Furthermore, the fault tolerance of the system is improved, and backup between management machines becomes possible.
[0033]
Here, in the data collection method according to the present embodiment, two intermediate units (SU) are paired so that they can be backed up with each other. However, three or more units can be used in one set. Further, the number of terminal units (DU) belonging to the first collection range and the number of terminal units (DU) belonging to the second collection range are not limited to 50 and can be freely set.
[0034]
In the data collection method according to the present embodiment, data in the first collection range and the second collection range is always collected. However, the present invention is not limited to this method, and only when one intermediate unit fails. The intermediate collection unit may collect data in the second collection range.
In the data collection method according to the present embodiment, the data in the second collection range is collected at 1/10 the frequency of the data in the first collection range, but the priority of data collection can be changed as appropriate.
In the data collection method according to the present embodiment, only the data in the first collection range is transmitted from the intermediate unit to the management device, but the data in the first collection range and the second collection range is transmitted. Is also possible.
[0035]
【The invention's effect】
According to the present invention, even if one intermediate unit fails, the data of the terminal units within the range can be collected by the management machine via another intermediate unit. For this reason, the fault tolerance of the system is improved.
[Brief description of the drawings]
FIG. 1 is a main part wiring block diagram of a data collection system for a pachinko machine according to an embodiment of the present invention.
FIG. 2 is a wiring block diagram of a conventional data collection system for a pachinko machine.
[Explanation of symbols]
DU terminal unit SU Intermediate unit P pachinko machine

Claims (4)

In a data collection method for collecting data of a plurality of terminal units by a management machine via a plurality of intermediate units,
Connecting the terminal unit and the intermediate unit with a communication line;
Collect data for each terminal unit with different priorities by at least two intermediate units ,
The intermediate unit is
For a plurality of connected terminal units, it is divided into a terminal unit that collects high priority data and a terminal unit that collects low priority data, and the data of each terminal unit with each divided priority Collect and
A data collection method for transmitting data of each terminal unit collected at different priorities to a management machine . The data collection method according to claim 1,
Normally, data collected by a high-priority intermediate unit is collected by the management machine, and when the high-priority intermediate unit fails, data collected by the low-priority intermediate unit is collected by the management machine. Method. In the data collection method according to claim 1 or 2,
A data collection method for connecting the intermediate unit and the management device via a communication line different from a communication line connecting the terminal unit and the intermediate unit. The data collection method according to any one of claims 1 to 3,
Divide multiple terminal units into n collection ranges,
Collecting data of terminal units within two collection ranges with one pair of intermediate units by pairing multiple intermediate units into two pairs,
One intermediate unit in the two intermediate units constituting the pair collects data at a high priority frequency from the terminal units in one collection range in the two collection ranges, and the high priority from the terminal unit in the other collection range Collect data at a low priority frequency that is lower than the frequency,
The other intermediate unit in the two intermediate units constituting the pair collects data at a high priority frequency from the terminal unit in the other collection range in the two collection ranges, and the high priority from the terminal unit in the one collection range. Collect data at a low priority frequency that is lower than the frequency,
The management machine
Collect the data of the terminal unit in the collection range collected at the high priority frequency from each intermediate unit, and collect the data from the terminal unit in the collection range that should be collected by the failed intermediate unit at the high priority frequency. Backup with data from the terminal unit in the collection range collected with low priority frequency from the intermediate unit that is paired with the intermediate unit
A data collection method characterized by the above.

Images