JP3600985B2 - How to extract specific product data - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method for extracting specific product data from among a plurality of product data stored in a database, which extracts specific product data having a high access count or distribution request or product data having low inventory (remaining amount). It is.
[0002]
[Prior art]
For the extraction method of specific product data that extracts specific product data with a high access count or distribution request or product data with a small inventory amount (remaining amount) from among a plurality of product data stored in the database, the sorting process is used. A database system for subscriber information of a telephone network is known for specifying data that is frequently used but frequently accessed in a database.
An intelligent network composed of three layers, a service operation layer, a service control layer, and a service management layer, separates an advanced communication service function and a related network control function hierarchically from an information transmission network such as a telephone network. Subscriber information, such as the conditions under which subscribers receive what network services, is managed in a service control layer database, and the service control layer database is configured as a distributed database system. General.
[0003]
In a distributed database system that constitutes a service control layer of an intelligent network, in deciding which database system to distribute subscriber information to be distributed, use of a network service provided to each subscriber. In anticipation of the situation, data is arranged in consideration of the load balance so that the load is not concentrated on a specific distributed database system.
[0004]
However, in the arrangement of subscriber information in anticipation of the usage status of the network service, when an unexpected usage status occurs, one or a plurality of specific subscribers located in a specific distributed database system may be allocated. In the case where service requests are concentrated at the same time, the load is concentrated on the specific distributed database system and the processing capacity is squeezed, so that data reference is delayed and normal network service cannot be received. May occur.
[0005]
To solve this problem, Japanese Patent Application Laid-Open No. Hei 5-12338 discloses a known example in which the number of data references and the average reference time for each database are collected during operation, and the data arrangement of the database is changed based on the collected information. .
[0006]
Japanese Patent Application Laid-Open No. 6-119383 discloses a known example in which frequently used data is multiplexed and arranged in a plurality of frequently used databases to minimize reference delay time.
[0007]
As a known example of a method of specifying data having a high access frequency, such as a data access frequency or a use frequency in these known examples, there is a method disclosed in Japanese Patent Application Laid-Open No. 1-177673.
[0008]
FIG. 19 is a diagram showing a record format of a frequency table for realizing the method disclosed in Japanese Patent Laid-Open No. 1-177673, in which a first field 1101 is a node id and corresponds to each data name of the database. Each of the second field 1102, the third field 1103, the fourth field 1104, and the fifth field 1105 is the number of accesses within a certain period, for example, in units of one week, the second field is this week, and the third field is last week. .., The week unit becomes older, and the sixth field 1106 indicates the total number of accesses in the second to fifth fields 1102 to 1105.
[0009]
As a method of specifying the node id having a high access frequency from the frequency table configured as described above, the records of the frequency table are sorted in the order of the total number of accesses (sixth field 1106), and a certain threshold is applied to the sorted records. Discloses a method of selecting a node id having a high access frequency in the above.
[0010]
[Problems to be solved by the invention]
However, according to the method of specifying a node id having a high access frequency disclosed in the above-mentioned Japanese Patent Application Laid-Open No. 1-177673, it is necessary to sort the records of the frequency table in the order of the total number of access times 1106. When the number of records becomes large, that is, the number of node ids 1101 increases, the processing required for sorting increases.
[0011]
The number of subscribers who receive network services using the intelligent network is enormous, and the number of subscribers accommodated in one distributed database system is tens of thousands to hundreds of thousands.
[0012]
For this reason, if the above-mentioned known example is applied to a distributed database system of an intelligent network to specify data having a high access frequency, it is necessary to perform sort processing on tens of thousands to hundreds of thousands of data. As a result, there is a problem that the influence on the processing capacity of the distributed database system is increased, and the original database access processing is squeezed.
[0013]
In this case, as a method of extracting data having a high access frequency without sorting, there is a method of registering only data in which the number of accesses exceeds a certain threshold. If it becomes full quickly and it is necessary to reset the threshold value, or if the threshold value is set too large, it takes time until data exceeding the threshold value is generated. When the data is frequently concentrated, the data that is frequently accessed, which is the cause of the load concentration, is to be relocated to a distributed database system with a small load. Takes a long time.
Such a problem arises in a system for realizing online shopping, for example, when specifying which product is accessed frequently, which distribution request is large, or which product inventory is low. Happens as well.
[0014]
An object of the present invention is to provide a system including a database that stores product data, such as an online shopping system, in which sorting processing is unnecessary, and specific product data that is frequently accessed without setting a threshold value in advance. It is an object of the present invention to provide a method for extracting specific product data that can quickly extract specific product data with a large distribution request or specific products with a small number of stocks.
[0015]
[Means for Solving the Problems]
In order to achieve the above object, the present invention provides a method for storing specific product data having a high access frequency among a plurality of product data stored in a database.Depending on the database systemA method for extracting specific product data to be extracted,The database system is
PredeterminedA first step of setting a threshold “0” for determining the number of accesses to the product data in each cycle in the threshold storage unit;Of the databaseeachMeasure the number of accesses to product data for each product data nameSecondSteps and the measured number of accessesSet in the threshold storage unitDetermine if a threshold has been exceededThirdSteps andThirdIf the result of the step is that the save area has been exceeded, the corresponding product data name is stored in the save area, and then it is determined whether or not the save area is full.4thAnd the current number of accesses of any one of the threshold-exceeding product data names registered in the save area if the storage area is full as a new thresholdSetting the save area to a state where the product data name is not stored in the threshold storage unit, and performing the third to fifth steps within the predetermined period. Repeat for each access to theIt is characterized in that the product data of the product data name stored in the save area is extracted as specific product data having a high access frequency.
In addition, the value of the current number of times of access is the maximum value among two or more threshold-exceeding product data names registered in the save area,SaidTo set a new thresholdFeatures.
[0016]
In addition, product data of a specific product having a small remaining amount among the products represented by a plurality of product data stored in the database is stored.Depending on the database systemA method for extracting specific product data to be extracted,The database system is
A first step of setting, in a threshold value storage unit, a maximum value that can be expressed in a stock quantity area of product data as a threshold value for determining a remaining amount associated with an order for a product in a predetermined cycle;The remaining amount associated with the order for the product indicated by the product data in the database is measured for each product data nameSecondSteps and the measured remainingSet in the threshold storage unitDetermine if less than thresholdThirdSteps andThirdAs a result of the step, if the number is small, the corresponding product data name is stored in the save area, and then it is determined whether or not the save area is full.4thStep and, if full, the remaining amount of any one of the threshold dropping product data names registered in the save area as a new threshold valueSetting the save area to a state where the product data name is not stored in the threshold storage unit, and performing the third to fifth steps within the predetermined period. Repeat for each order receivedIt is characterized in that the product data of the product data name stored in the save area is extracted as specific product data having a small remaining amount.
Further, a value of the current remaining amount among the two or more threshold dropping product data names registered in the save area is set toSaidIt is characterized in that it is set as a new threshold.
[0017]
In addition, specific product data with a high distribution request among a plurality of product data stored in the databaseDepending on the database systemA method for extracting specific product data to be extracted,
The database system is
PredeterminedA first step of setting a threshold “0” for determining the number of distribution requests to the product data for each cycle in the threshold storage unit;Of the databaseeachMeasure the number of distribution requests to product data for each product data nameSecondSteps and the measured number of delivery requestsSet in the threshold storage unitDetermine if a threshold has been exceededThirdSteps andThirdIf the result of the step is that the save area has been exceeded, the corresponding product data name is stored in the save area, and then it is determined whether or not the save area is full.4thAnd the current number of distribution requests for any one of the threshold-exceeding product data names registered in the save area if the storage area is full as a new threshold.Setting the save area to a state where the product data name is not stored in the threshold storage unit, and performing the third to fifth steps within the predetermined period. Repeat for each delivery request toIt is characterized in that the product data of the product data name stored in the save area is extracted as specific product data having a high distribution request frequency.
In addition, among the two or more threshold-exceeding product data names registered in the save area,Delivery request countIs the largest valueThe aboveIt is characterized in that it is set as a new threshold.
[0018]
According to the above means, while dynamically updating the threshold value, the commodity data with a high access frequency, the commodity data with a large distribution request, or the commodity data with a small stock quantity are narrowed down to the save area over time.
For this reason, when extracting the product data with a high access frequency, the product data with a high distribution request, or the product data with a small number of stocks, it is not necessary to search all the product data and sort them in the order of the access frequency. The effect on the processing capacity of the system can be reduced.
[0019]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, the present invention will be described in detail based on an illustrated embodiment.
[0020]
FIG. 1 is a block diagram of an embodiment in which the present invention is applied to a case where data frequently accessed in a distributed database in an intelligent network is specified.
[0021]
In FIG. 1, a distributed database system management unit 101 corresponds to a service management layer 190 in an intelligent network, and is connected to a plurality of distributed database systems 102 via a communication network 103.
[0022]
The distributed database system 102 corresponds to the service control layer 191 in the intelligent network, and has a database access unit 105 that controls access to the database 106 accommodated in its own system, and an access frequency collection unit that collects the access frequency to the database 106. 108, a communication control unit 104 for connecting the distributed database system management unit 101 or another distributed database system 102 with the communication network 103, and processing of the communication control unit 104, the database access unit 105, and the access frequency collection unit 108. It comprises a CPU usage rate measuring unit 107 for collecting the usage rate of a central processing unit (hereinafter abbreviated as CPU) to be executed.
[0023]
Here, the CPU usage rate is a ratio of CPU time allocated to communication control, database access, and access frequency collection per unit time.
[0024]
The distributed database system management unit 101 of the present embodiment is the same as the conventional configuration, and detailed description of its configuration and operation is omitted. However, basically, the distributed database system The CPU usage rate and the name of the data with high access frequency at that time are received from each distributed database system 102, and based on the information, the distributed database system 102 with the low CPU usage rate has a higher CPU usage rate. In this method, the arrangement of frequently accessed data arranged in the distributed database system 102 is changed.
[0025]
The access frequency collection unit 108 in each of the distributed database systems 102 collects the CPU usage rate required for this and the data names with high access frequency.
[0026]
The access frequency collection unit 108 includes an access count addition unit 109 that counts the number of accesses to each data, a threshold excess information storage unit 110 that stores data that exceeds a threshold set by the dynamic threshold change unit 111, and a threshold excess When the data name save area becomes full, the dynamic threshold changing unit 111 changes the threshold again and determines a new threshold for collecting frequently accessed data again. And a threshold excess information periodic transmission unit 112 for notifying the CPU usage rate and the name of the frequently accessed data at that time, and performing processing for collecting the frequently accessed data again after completion of the notification.
[0027]
The threshold excess information storage unit 110 is an area for saving various kinds of information necessary to store the threshold information that changes every moment or the name of the data that exceeds the threshold, and the threshold measurement information 201 stored here is shown in FIG. As shown, it is composed of an access frequency collection identifier 202, a threshold 203, a threshold excess identifier 204, and a save area identifier 205.
[0028]
Here, the access frequency collection identifier 202 indicates that when the access frequency collection for a certain period of time is completed, the notification to the distributed database system management unit 101 is completed, and the collection of the access frequency is newly started, This is the information for distinguishing the collection from the collection.
[0029]
The threshold value 203 is a current threshold value that is updated by the dynamic threshold value change unit 111 each time the threshold excess data name save area 401 in FIG.
[0030]
The threshold excess identifier 204 is information for identifying whether or not the number of accesses of each data exceeds the threshold 203, whether or not the data has already been saved in the threshold excess data name save area 401 shown in FIG.
[0031]
The save area identifier 205 is information for identifying which of the threshold-exceeding data name save area 401 composed of the side A and the side B shown in FIG. 4 is currently used.
[0032]
FIG. 3 shows the configuration of each piece of data arranged in the database 106, including a data name 301, an access frequency collection identifier save 302 for saving an access frequency collection identifier 202 of the threshold measurement information 201, and a threshold of the threshold measurement information 201. It comprises a threshold excess identifier save 303 for saving the excess identifier 204, an access count 304 for saving the access count, and actual data 305.
[0033]
FIG. 4 shows an area for storing the data name when the number of accesses 304 exceeds the threshold value 203. The data area save data area 401 with the excess threshold value has a two-sided configuration of A side and B side. A threshold save 402 for saving the threshold 203 of the information 201, a registered number 403 indicating the number registered in the threshold excess data name save area 401, and a finite number (n in FIG. 4) of data names are stored. And a threshold excess data name 404.
[0034]
The processing outline of the access count adding unit 109 of the access frequency collecting unit 108 will be described with reference to FIG.
[0035]
When an access request to the database 106 occurs, control is passed from the database access unit 105 to the access count addition unit 109. The access count adding unit 109 extracts the access frequency collection identifier save 302 of the relevant data from the database 106 (step 501).
[0036]
It is determined whether or not the value of the extracted access frequency collection identifier save 302 matches the access frequency collection identifier 202 of the threshold measurement information 201 (step 502). In the case of a mismatch, the access frequency collection identifier 202 of the threshold measurement information 201 is set in the access frequency collection identifier save 302 of the relevant data because it is the first database access request since the access frequency collection is newly started (step 504). ), "0" is set to the access count 304 of the data (step 505), and "1" is added to the access count 304 of the data (step 506).
[0037]
As a result, “1” is set to the number of accesses 304 as the first database access request after newly starting the access frequency collection.
[0038]
On the other hand, if it is determined in step 502 that there is a match, since the access data has been accessed at least once since the access frequency was newly collected, “1” is added to the access count 304 of the relevant data. (Step 506), the latest access count 304 is set.
[0039]
Finally, control is passed from the access count adding unit 109 to the threshold excess information storage unit 110 (step 508).
[0040]
The processing outline of the threshold excess information storage unit 110 of the access frequency collection unit 108 will be described with reference to FIG.
[0041]
When the control is passed from the access count adding unit 109, the threshold excess information storage unit 110 extracts the access count 304 of the corresponding data calculated by the access count adding unit 109 (step 601). A comparison is made with the threshold 203 of the threshold measurement information 201 (step 602).
[0042]
If the number of accesses 304 exceeds the threshold 203, a threshold excess identifier save 303 of the relevant data is extracted to determine whether or not it has already been registered in the threshold excess data name save area 401 due to the current threshold excess (see FIG. Step 604), it is determined whether or not the extracted threshold excess identifier save 303 matches the threshold excess identifier 204 of the threshold measurement information 201 (step 605).
[0043]
If the threshold excess identifier save 303 and the threshold excess identifier 204 of the threshold measurement information 201 do not match, it is the first database access request after the change to the new threshold, so the threshold excess data name save area corresponding to the save area identifier 205 The data name 301 of the corresponding data is registered in 401, the registration number 403 is incremented by “1” (Step 607), and the threshold excess identifier 204 of the threshold measurement information 201 is set in the threshold excess identifier save 304 of the data. (Step 608).
[0044]
Thereafter, control is passed to the dynamic threshold changing unit 111 (step 609).
[0045]
On the other hand, if it is determined in step 602 that the threshold value has not been exceeded, and if it is determined in step 605 that the data has already been registered in the threshold-exceeding data name save area 401, and the match has been determined, the processing of the access frequency collection unit 108 ends ( Step 612).
[0046]
The processing outline of the dynamic threshold value changing unit 111 of the access frequency collecting unit 108 will be described with reference to FIG.
[0047]
When the control is passed from the threshold excess information storage unit 110, the dynamic threshold change unit 111 determines whether the currently used threshold excess data name save area 401 is full of the registration area (step 701).
[0048]
When the registration area is full, any one threshold excess data name 404 registered in the excess threshold data name save area 401 is extracted (step 703), and the current access count 304 of the extracted data name is set to a new value. The threshold is set to the threshold 203 of the threshold measurement information 201 (step 704).
[0049]
The method of arbitrarily selecting in the step 703 is that the excess threshold data name (1) 404 registered at the top of the excess threshold data name save area 401 or the i-th registered excess threshold data name (i) 404 is used. There may be.
[0050]
The arbitrarily selected data is registered again no matter how many times there is a data access request until the threshold excess data name save area 401 becomes full and the dynamic threshold changing unit 111 updates the threshold. This is not the case, and only the access count 304 is updated. For this reason, the access count 304 of the data at the time selected in step 703 is larger than the current threshold.
[0051]
Therefore, if the data arbitrarily selected in step 703 is data with a high access frequency, the updated threshold is set to a value larger than the current threshold. A value larger by "1" than the current threshold value 203 is set, and data with a high access frequency is narrowed down using the new threshold value.
[0052]
Next, the threshold excess identifier 204 is updated (step 705), and the save area identifier 205 is changed (step 706). Finally, the threshold of the threshold excess data name save area 401 corresponding to the changed save area identifier 205 is changed. The new threshold value 203 is set in the save 402, and “0” is set in the registration number 403 (step 707).
[0053]
On the other hand, if the registration area is not full at step 701, the threshold change is unnecessary, and the dynamic threshold change process ends.
[0054]
The threshold excess information periodical transmission unit 112 of the access frequency collection unit 108 arbitrarily pre-sets the threshold excess data name 404 registered in the threshold excess data name save area 401 collected by the threshold excess information storage unit 110 in the system. A function of notifying the distributed database system management unit 101 at a set fixed time interval is realized.
[0055]
The processing outline of the threshold excess information periodic transmission unit 112 will be described with reference to FIGS.
[0056]
8 shows the configuration of information to be transmitted to the distributed database system management unit 101, and shows the average within a certain time period measured by the CPU usage rate measurement unit 107 using the conventional technique. The area includes a CPU usage rate 802 which is an area for storing the CPU usage rate, a data name 804 exceeding the threshold, and an access count 805 for storing the access count for the data at the time of transmission.
[0057]
The threshold excess information periodic transmission unit 112 collects the CPU utilization from the CPU utilization measurement unit 107, sets the CPU utilization 802 of the threshold excess data name transmission information 801 (step 901), and saves the threshold excess data name save area (step 901). The current number of accesses 304 of the data name 404 registered in the data area 404 registered in the data area 404 registered in the threshold excess data name save area (side B) 401 is collected, and the data name 804, the access number 805, and the number of transmissions 803 are set. (Steps 902 and 903).
[0058]
Next, “0” is set to the registration number 403 on both sides of the threshold excess data name save area 401 (step 904), the access frequency collection identifier 202 of the threshold measurement information 201 is updated (step 905), and the threshold measurement information is further updated. The threshold 203 of 201 is initially set to “0” (step 906), and the process ends.
[0059]
As a result, the collection of frequently accessed data at fixed time intervals is restarted.
[0060]
A specific example in which the threshold value is dynamically changed will be described with reference to FIG.
In the specific example of FIG. 10, the number of data that can be registered in the threshold excess data name save area 401 is set to five, and the data is already registered in the save area, and the remaining one data is used as the save area A Indicates that the registration area 401 is full and the threshold is updated.
[0061]
When a database access request for the data name A5 occurs, the processing of the access count adding unit 109 is performed.
[0062]
Then, since the access frequency collection identifier save 1004 of the data name A5 matches the access frequency collection identifier 1001 of the threshold measurement information 201, the access count 1006 of the data name A5 is added by the processing of step 506 in FIG.
[0063]
In the threshold excess information storage unit 110, since the access count 1017 of the data name A5 is larger than the threshold 1002 of the threshold measurement information 201, the process of step 604 in FIG. 6 is executed. Also, since the threshold excess identifier save 1005 of the data name A5 does not match the threshold excess identifier 1003 of the threshold measurement information 201, the processing of step 607 in FIG. 6 is executed, and the threshold of the A side currently used as the save area is executed. The data name A5 is registered in the excess data name (5) 1012 of the excess data name save area 401, and the registered number 1010 is updated.
[0064]
Next, the threshold excess identifier 1003 of the threshold measurement information 201 is set in the threshold excess identifier save 1016 of the corresponding data A5, and the control is passed to the dynamic threshold changing unit 111.
[0065]
Since the registration area is full, the dynamic threshold changing unit 111 executes the processing after step 703 in FIG. 7 and sets a new threshold. As an example of a method for extracting an arbitrary threshold-exceeding data name in step 703 in FIG. 7, FIG. 10 shows a case where a threshold-exceeding data name registered at the head of a registration area is extracted as an example.
[0066]
In the example of FIG. 10, the data name A1 is registered at the top 1011 of the A side of the currently used threshold excess data name save area 401, and the threshold measurement is performed using the current access count 1015 of the data A1 as a new threshold. The threshold value 1007 of the information 201 is set.
[0067]
Next, the threshold excess identifier 1008 is updated, and the save area identifier 1009 is changed to the B side.
Finally, a new threshold value 1007 is set in the threshold save 1013 of the threshold excess data name save area 401 on the B side, and the registration number 1014 is initialized to “0”, and the processing of the dynamic threshold change unit 111 ends.
[0068]
In the processing example of the dynamic threshold updating unit 111 illustrated in FIG. 7, a method of extracting an arbitrary one of the threshold excess data names 404 (Step 703) and setting the access count 304 of the data at that time as the new threshold 203. However, among the plurality of threshold-exceeding data names 404, the one with the largest number of accesses 304 at that time may be used as the new threshold 203. According to this method, the new threshold value is updated based on the number of accesses of the data having a higher access frequency, so that the data having the higher access frequency can be narrowed down more quickly, and the data having the higher access frequency can be identified. If you hurry, it will be more effective.
[0069]
Further, in the processing example of the threshold excess information periodic transmission unit 112 illustrated in FIGS. 8 and 9, the access count 304 at that time of the threshold excess data name 404 registered in the threshold excess data name save area 401 is extracted. Then, the method of transmitting the data to the distributed database system management unit 101 has been described, but the two threshold excess data name save areas 401 shown in FIG. 4 may be transmitted as they are.
[0070]
When the initial setting of the threshold value 203 is performed (step 906), if the initial value is set not to “0” but to a condition such as “現在” of the current threshold value 203, the access performed after the first periodic transmission is performed. Frequent data can be narrowed down more quickly.
[0071]
Further, the distributed database system 102 does not autonomously identify frequently accessed data and does not periodically transmit the result. Instead, the distributed database system management unit 101 identifies the frequently accessed data according to an instruction from the distributed database system management unit 101. Alternatively, a method of starting and reporting the result according to an instruction from the distributed database system management unit 101 may be adopted.
[0072]
FIG. 11 is a diagram showing a configuration of a system in which a catalog seller realizes an online home shopping service, so-called virtual shopping.
The catalog distributor 1201 processes, on the server 1202, a request for more detailed information on the product described in the catalog or an order for the product, which is requested from each client 1206 of the customer 1205 via the network 1204, and processes the request from the server 1202. It provides the merchandise information stored in 1203 to the client 1206 or updates the stock status accompanying the order from the client 1206.
[0073]
In virtual shopping, it is difficult to accurately predict the number of accesses for each product, such as which product accesses more detailed information of which product, or the number of orders received for each product. For this reason, if it is possible to identify the products that are frequently accessed by customers and to change the information provided as appropriate, such as to make more detailed information on the products that are frequently accessed or to lower the prices of those products, it is possible to receive orders. It is also possible to further increase.
[0074]
In addition, if a case where orders are concentrated and inventory becomes insufficient can be detected at an earlier stage, it is not necessary to refuse orders or delay delivery of products, and it is possible to further improve serviceability to customers.
[0075]
However, as the number of target products increases, the number of frequently accessed products and orders increases, and when it is attempted to specify a low-stock product in real time, the effect on the processing capacity of the server 1202 becomes large, and the customer 1205 is sent to the customer 1205. The response is delayed.
[0076]
Therefore, by applying the present invention, it is possible to realize a system for specifying a product with a high access frequency in real time and specifying a product with a small stock. FIG. 12 shows the configuration of the threshold value measurement information 201 for realizing such a system, FIG. 13 shows the configuration of the database 1203, and FIG. 14 shows the configuration of the area for saving products with low stock.
[0077]
In FIG. 12, 202 to 205 are the same information as described in FIG. In the present embodiment, a stock threshold 1301, a stock threshold drop identifier 1302, and a stock threshold drop save area identifier 1303 are newly provided. These pieces of information are used when identifying frequently accessed data. Corresponding to the threshold 203, the threshold excess identifier 204, and the save area identifier 205.
[0078]
In FIG. 13, the stock number 1402 is data that is decremented as needed upon receiving an order from the client 1206, and the stock threshold drop identifier save 1401 indicates that the stock quantity 1402 of each product is smaller than the stock threshold 1301 when the stock quantity 1402 is smaller than the stock threshold 1301. This is information for identifying whether or not the data has already been saved in the threshold drop data name save area 1501.
[0079]
It should be noted that the specification of a product with a high access frequency can be realized by the processing described in FIGS. 5, 6, 7, and 9. However, the information on the CPU usage rate described in FIG. 9 is unnecessary.
[0080]
On the other hand, in order to specify a product with a small stock, the stock threshold is dynamically reduced. When the stock threshold drop data name save area 1501 shown in FIG. 14 becomes full, In the same flow as the processing flow of the dynamic threshold changing unit 111 shown in FIG. 7, a new inventory threshold 1301 is dynamically determined based on the current inventory quantity of the data arbitrarily selected from the inventory threshold drop data name save area 1501. Will be updated.
[0081]
That is, in the embodiment of FIG. 1, the threshold value is increased in accordance with the access frequency. However, as in the case of specifying a product with a small stock, the number, the number, the ratio, the required time, and the like of the identification target can be measured. When specifying a physical quantity that is small, the threshold value is dynamically updated in a smaller direction.
[0082]
As the initial value of the stock threshold 1301, the maximum value that can be represented by the area (byte length / bit length) of the stock quantity 1402 in the database 1203 is used. Alternatively, if the appropriate stock quantity is known in advance for each product, the ratio (%) of the stock quantity to the proper stock quantity can be set as the stock threshold instead of the stock quantity. A value of “100 (%)” may be used as the initial value of the threshold value 1301.
[0083]
In the system configuration of FIG. 11, by executing the processing described in FIGS. 5, 6, 7, and 9 on the server 1202, the virtual shopping does not impose a large burden on the server 1202, and the customer can select which merchandise item. It is possible to accurately measure the number of accesses for each product, such as whether more detailed information is accessed, or the number of orders for each product. As a result, it is possible to contribute to an increase in the number of orders by making the information of the products that are frequently accessed more detailed or reducing the prices of the products.
[0084]
In addition, it is possible to identify products that have a shortage of inventory due to a concentration of orders at an early stage, and to avoid rejecting orders or delaying the delivery of products, thereby further improving serviceability to customers.
[0085]
FIG. 15 shows a video-on-demand service using a video server having a distributed processing configuration having a load distribution function when access to specific video software (hereinafter abbreviated as video stream) is concentrated. FIG. 1 is a diagram illustrating a configuration of a system that provides a VOD service (hereinafter, abbreviated as VOD service).
[0086]
The VOD service provider 1601 receives a request from each client 1610 of the customer 1609 at the video server 1602 via the network 1608, and distributes a video stream to the client. In the example of FIG. 15, the video server 1602 comprises three load-balanced video stream transmission units 1604 and one VOD management unit 1606, each of which has a video stream storage unit 1603 and an access frequency. An access frequency database 1605 for collection is provided, and the VOD management unit 1606 and the video stream transmission unit 1604 are connected by a network 1607.
[0087]
In the video server 1602 having a load distribution configuration, when a case where requests are concentrated on a specific video stream occurs, the video stream on which requests are concentrated is converted to the video stream of the video stream transmitting unit 1604 having the lightest load. It has a function of distributing the load of the video stream transmitting unit 1604 by multiplexing and arranging it in the stream storage unit 1603 so as to avoid a situation in which a request from the client 1610 cannot be satisfied in advance.
[0088]
In the case of the VOD service as well, the distribution of demand from customers is not uniform, changes every moment, and depending on social factors, there are cases where a particular video stream is concentrated on a specific date and time. The more it becomes, the more difficult it is to identify frequently accessed video streams in real time.
[0089]
Thus, the present invention is applied to a video server having such characteristics, and the structure of threshold measurement information when specifying a video stream with high access frequency is shown in FIG. FIG. 17 shows the structure of the access frequency database 1605 in FIG.
[0090]
The threshold measurement information 1701 in FIG. 16 includes, for each of the three video stream transmission units 1604, a threshold 203, a threshold excess identifier 204, a save area identifier 205, and a video stream transmitted by the video stream transmission unit 1604. It comprises a total number of transmission streams 1703 for counting the total number.
[0091]
The threshold excess data name save area 1801 composed of two sides A and B is also configured to correspond to each of the three video stream transmission units 1604.
The access frequency database 1605 managed by the VOD management unit 1606 has a configuration in which the threshold excess identifier save 303 and the number of accesses 304 correspond to each of the three video stream transmission units 1604.
[0092]
The specification of a video stream with a high access frequency is realized by executing the processing described in FIGS. 5, 6, 7, and 9 for each of the three video stream transmission units.
[0093]
The multiplex arrangement of the video streams is such that the video stream transmitting section 1604 having the smallest total number of transmission streams 1604 has the data name save area exceeding the threshold specified by the video stream transmitting section 1604 having the largest total number of transmission streams 1703. This can be realized by multiplexing the video streams with the highest access frequency registered in 1801.
[0094]
As a result, it is possible to quickly cope with a case where access to a specific video stream is concentrated on a specific date and time, and to improve service for a user.
[0095]
Note that the present invention is not limited to the above-described embodiment, and is applied to all cases in which a measurable physical quantity such as the number, the number, the ratio, and the required time of a specific object is small or large. be able to.
[0096]
【The invention's effect】
As described above, according to the present invention, in a system provided with a database storing product data, such as an online shopping system, a sorting process is unnecessary, and a high frequency of access can be specified without setting a threshold value in advance. It is possible to quickly extract product data, specific product data with a large distribution request, or specific products with a small number of stocks, and reduce the processing capacity of the system. The extracted merchandise data can be used as reference for expanding and rearranging the merchandise data, supplementing the stock amount, and the like.
[Brief description of the drawings]
FIG. 1 is a block diagram of a distributed database system according to an embodiment of the present invention.
FIG. 2 is a diagram showing a configuration of threshold measurement information in FIG. 1;
FIG. 3 is a diagram showing a configuration of a database in FIG. 1;
FIG. 4 is a diagram showing a configuration of an area for saving a threshold excess data name in FIG. 1;
FIG. 5 is a flowchart showing a processing procedure of an access number adding unit in FIG. 1;
FIG. 6 is a flowchart illustrating a processing procedure of a threshold excess information storage unit in FIG. 1;
FIG. 7 is a flowchart illustrating a processing procedure of a dynamic threshold value changing unit in FIG. 1;
FIG. 8 is a diagram illustrating a configuration of information notified to a distributed database system management unit in FIG. 1;
9 is a flowchart illustrating a processing procedure of a threshold excess information periodic transmission unit in FIG. 1;
FIG. 10 is an explanatory diagram showing a specific example of dynamic threshold change according to the present invention.
FIG. 11 is a block diagram showing an embodiment of an online shopping service system to which the present invention is applied.
FIG. 12 is a diagram illustrating a configuration of threshold measurement information in FIG. 11;
FIG. 13 is a diagram showing a configuration of a database in FIG. 11;
FIG. 14 is a diagram showing a configuration of an area for saving a data name exceeding a threshold in FIG. 11;
FIG. 15 is a block diagram showing an embodiment of a video-on-demand service system to which the present invention is applied.
FIG. 16 is a diagram showing a configuration of threshold measurement information in FIG.
FIG. 17 is a diagram showing a configuration of an area for saving a threshold excess data name in FIG. 15;
18 is a diagram showing a configuration of an access frequency database in FIG.
FIG. 19 is a diagram showing a configuration of a conventional access frequency table.
[Explanation of symbols]
108: access frequency collection unit, 109: access count addition unit, 110: threshold excess information storage unit, 111: dynamic threshold change unit, 112: threshold excess information periodic transmission unit, 203: threshold, 304: access count, 401 ... Threshold excess data name save area.

Claims (6)

An extraction method of specific product data, wherein a specific product data having a high access frequency among a plurality of product data stored in a database is extracted by a database system ,
The database system is
A first step of setting a threshold “0” for determining the number of accesses to the product data at predetermined intervals in a threshold storage unit;
A second step of measuring the number of accesses to each product data of the database within the predetermined cycle for each product data name;
A third step of determining whether the measured number of accesses has exceeded a threshold set in the threshold storage unit ,
If the result of the third step is that the product data name is exceeded, the corresponding product data name is stored in the save area, and then a fourth step of determining whether or not the save area is full;
If it is full, the current access count of any one of the product data names exceeding the threshold value registered in the save area is set as a new threshold value in the threshold storage unit, and the save area does not store the product data name. And a fifth step of setting the state
The third to fifth steps are repeated for each access to the product data within the predetermined period, and the product data of the product data name stored in the save area is extracted as specific product data having a high access frequency. A method for extracting specific product data. 2. The specific product data according to claim 1, wherein a value having the largest current access count among two or more product data names exceeding the threshold value registered in the save area is set as the new threshold value. 3. Extraction method. A method of extracting specific product data by extracting, using a database system, product data of a specific product having a small remaining amount among products indicated by a plurality of product data stored in a database ,
The database system is
A first step of setting, in a threshold storage unit, a maximum value that can be expressed in a stock quantity area of product data as a threshold for determining a remaining amount associated with an order for a product at predetermined intervals;
A second step of measuring a remaining amount associated with an order for a product indicated by the product data in the database within the predetermined cycle, in units of product data name;
A third step of determining whether or not the measured remaining amount is smaller than a threshold set in the threshold storage unit ;
If the result of the third step is that there is less, the corresponding product data name is stored in the save area, and then a fourth step of determining whether or not the save area is full;
If the storage area is full, the remaining amount of the commodity of any one of the threshold-dropped commodity data names registered in the save area is set as a new threshold in the threshold storage unit, and the save area does not store the commodity data name. And a fifth step of
The third step to the fifth step are repeated every time an order is received for the product data within the predetermined period, and the product data of the product data name stored in the save area is extracted as specific product data having a small remaining amount. A method for extracting specific product data. 4. The specific product data according to claim 3, wherein a value with the smallest remaining amount among two or more product names of threshold-drop products registered in the save area is set as the new threshold. 5. Extraction method. A specific product data extraction method for extracting a specific product data having a large distribution request among a plurality of product data stored in a database by a database system ,
The database system is
A first step of setting a threshold “0” in the threshold storage unit for determining the number of distribution requests to the product data at predetermined intervals;
A second step of measuring, for each product data name, the number of distribution requests to each product data in the database within the predetermined cycle ;
A third step of determining whether the measured number of distribution requests has exceeded a threshold set in the threshold storage unit ,
If the result of the third step is that the product data name is exceeded, the corresponding product data name is stored in the save area, and then a fourth step of determining whether or not the save area is full;
If it is full, the current number of distribution requests for any one of the product data names exceeding the threshold value registered in the save area is set as a new threshold value in the threshold storage unit, and the save area stores the product data name. And a fifth step of eliminating the
The third step to the fifth step are repeated for each distribution request to the commodity data within the predetermined period, and the commodity data of the commodity data name stored in the save area is replaced with specific commodity data having a high distribution request frequency. A method for extracting specific product data, characterized in that the specific product data is extracted. 6. The specific product data according to claim 5, wherein a value having the maximum number of current distribution requests among two or more product data names exceeding the threshold registered in the save area is set as the new threshold. 7. Extraction method.

Images