JP6027504B2 - Application server and cache control method - Google Patents

Description

  The present invention relates to an application server that controls data in a cache, and a cache control method.

  By storing frequently accessed data in a cache, the time for reading data from a low-speed storage device is shortened to speed up the processing. For example, Non-Patent Document 1 describes storing presence data with high access frequency in a server as a cache.

Baba, et al., “IMS Presence Cache Method Based on Access Frequency”, IEICE Society Conference Proceedings 2010_Communications (2), p. 35

  In the method of performing caching in the order of high access frequency shown in Non-Patent Document 1, even when access is temporarily concentrated and there is almost no subsequent access, the priority order of cache remains high and there is almost no access. Regardless, there is a problem that cache data remains in the server.

  For example, when a server provides a billing service used for a company call center or the like, generally, the call center can accept a weekday daytime and cannot accept a nighttime or holiday. In such a case, the access frequency of the subscriber data providing the call center is high during weekdays and low during nights and holidays, but if the access frequency during weekdays is high, the access frequency is low at night and There is a problem that cache data remains on the server even on holidays.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to appropriately delete data whose access frequency is high in a certain period and thereafter decreases in frequency from the cache according to the use status of the data. An application server that effectively uses a cache and a cache control method are provided.

In order to solve the above problems, the present invention according to claim 1 is an application server, wherein a cache storage unit for storing data, and a priority for controlling the priority of each data stored in the cache storage unit A control unit, and a deletion unit that deletes data in descending order of priority when the capacity of the cache storage unit is insufficient, and the priority control unit includes priority levels of data stored in the cache storage unit. When the access to the data occurs, the priority of the data is increased, and the priority control means increases the speed of decreasing the priority with the passage of time outside the service providing time. In the service providing time, the speed is made faster than the speed at which the priority is lowered as time elapses.

  According to the second aspect of the present invention, in the application server, when access to the data occurs, the priority control unit calculates an increase value using the last access date and time and the current time of the data, Increase the priority by adding the increment value.

According to a third aspect of the present invention, in the application server, the data stored in the cache storage means is subscriber data, and the subscriber data used in the request received from the call control server is the cache storage. In the case where it does not exist in the means, it further comprises data acquisition means for reading out the subscriber data from the subscriber database and storing it in the cache storage means.

The present invention according to claim 4 is a cache control method performed by an application server including a cache storage unit, the priority control step of controlling the priority of each data stored in the cache storage unit, and the cache When the capacity of the storage unit is insufficient, a deletion step of deleting data in descending order of priority is performed, and the priority control step decreases the priority of the data stored in the cache storage unit as time passes In addition, when access to the data occurs, the priority of the data is increased, and the priority control step sets the speed of decreasing the priority as time elapses outside the service providing time within the service providing time. Make it faster than the rate of lowering priority over time.

6. The cache control method according to claim 5 , wherein when the access to the data occurs, the priority control step calculates an increase value using the last access date and time and the current time of the data. Then, the increment is added to increase the priority.

The present invention according to claim 6 is the cache control method, wherein the data stored in the cache storage unit is subscriber data, and the subscriber data used in a request received from a call control server is In the case where it does not exist in the cache storage unit, it further comprises a data acquisition step of reading the subscriber data from the subscriber database and storing it in the cache storage unit.

  According to the present invention, an application server that appropriately deletes data whose access frequency is high in a certain period and thereafter decreases in access frequency from the cache in accordance with the use state of the data and effectively uses the cache, and a cache control method Can be provided.

1 is an overall configuration diagram of a network system according to a first embodiment. It is a block diagram which shows the structure of an application server. It is a figure which shows an example of the data of a cache memory | storage part. It is a flowchart which shows the operation example of this embodiment. It is an image figure which shows the change of the priority of the data of a cache memory | storage part. It is a figure which shows an example of the data of the subscriber database of the Example of a receipt charge service. It is a sequence which shows the operation example in the service provision time of an Example. It is a sequence which shows the operation example outside the service provision time of an Example. It is an image figure which shows the change of the priority of the data of the cache memory | storage part of an Example.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram illustrating a configuration example of a network system according to an embodiment of the present invention. The network system of this embodiment is a system using SIP such as IMS (IP Multimedia Subsystem), for example, and includes an application server 1, a subscriber database 2, a plurality of call control servers 3, and each call control server. A plurality of terminals 4 connected to 3.

  The application server 1 provides a service such as an incoming billing service to the terminal 4. The application server 1 is connected to the subscriber database 2 and reads (downloads) subscriber data necessary for providing the service from the subscriber database 2 to provide the service. Further, when the application server 1 makes an inquiry to the subscriber database 2 at each service request, there are problems such as delay in session establishment and bandwidth compression due to an increase in communication frequency between the application server 1 and the subscriber database 2. For this reason, the application server 1 includes a cache storage unit and stores subscriber data read from the subscriber database 2 in the application server 1. The application server 1 corresponds to an AS (Application Server) in IMS.

  The subscriber database 2 stores the original subscriber data of each subscriber (user) who uses the terminal 4, and corresponds to an HSS (Home Subscriber Server) in IMS.

  The call control server 3 performs call control such as registration and authentication of each subordinate terminal 4 and establishment and control of a session. The call control server 3 corresponds to CFCS (call session control function) in IMS.

  The terminal 4 is a communication device used by a user such as a smartphone, a mobile phone, or a PC, for example.

  In the present embodiment, the following description will be given by taking as an example an embodiment in which the subscriber data of the subscriber database 2 is stored in the cache storage unit of the application server 1, but the data stored in the cache storage unit of the application server is The data is not limited to subscriber data, and may be other data.

  FIG. 2 is a block diagram showing the configuration of the application server 1 of the present embodiment. The illustrated application server 1 includes a service providing unit 11, a data acquisition unit 12, a priority control unit 13, and a cache storage unit 14 (cache memory).

  When the service providing unit 11 receives a request transmitted from the terminal 4 via the call control server 3, the service providing unit 11 performs the requested service processing. The data acquisition unit 12 (data acquisition unit, deletion unit) reads the subscriber data from the subscriber database 2 when the subscriber data used in the request received from the call control server 3 does not exist in the cache storage unit 14. And stored in the cache storage unit 14. In addition, when the capacity of the cache storage unit 14 is insufficient, the data acquisition unit 12 deletes data in descending order of priority.

  The priority control unit 13 controls the priority of each data stored in the cache storage unit 14. Specifically, the priority control unit 13 decreases the priority of the data stored in the cache storage unit 14 as time passes, and increases the priority of the data when access to the data occurs. .

  The cache storage unit 14 stores subscriber data read from the subscriber database 2. FIG. 3 is a diagram illustrating a configuration example of subscriber data stored / stored in the cache storage unit 14. In the present embodiment, the priority and the last use date are added to the subscriber data stored in the cache storage unit 14. The priority indicates the degree to which data stored in the cache storage unit 14 remains preferentially not deleted. The last use date and time indicates the date and time when the cache data was used last (immediately before).

  Note that the application server 1 described above can use a general-purpose computer system including, for example, a CPU, a memory, and an external storage device such as an HDD. In this computer system, each function of the application server 1 is realized by the CPU executing the program of the application server 1 loaded on the memory. The program for the application server 1 can be stored in a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM, an MO, a DVD-ROM, or can be distributed via a network.

  Next, the processing of this embodiment will be described.

  FIG. 4 is a flowchart showing the priority data priority setting process stored in the cache storage unit 14 of this embodiment.

  The service providing unit 11 of the application server 1 receives a request (for example, a new INVITE signal) transmitted from the terminal 4 via the call control server 3 (S11). The service providing unit 11 sends an acquisition instruction for the subscriber data used in the received request to the data acquiring unit 12. Thereby, the data acquisition unit 12 determines whether the instructed subscriber data exists in the cache storage unit 14 of the application server 1 (S12).

  When the designated subscriber data exists in the cache storage unit 14 (S13: YES), the data acquisition unit 12 acquires the subscriber data stored in the cache storage unit 14 (S14). And the service provision part 11 implements the service process provided to the terminal 4 using the subscriber data which the data acquisition part 12 acquired.

  Then, the priority control unit 13 updates the value of the priority P added to the subscriber data stored in the cache storage unit 14 (S15). Here, the priority control unit 13 adds the increase value Q to the priority P stored in the cache storage unit 14 to increase the priority P (P + Q → P). The increase value Q added to the priority P may be a constant or a value represented by the function g (s, t). The function g (s, t) is a function for calculating the increase value Q using the last use date and time s when the subscriber data in the cache storage unit 14 was last used and the current time t. A specific example of the function g (s, t) will be described later.

  Then, the priority control unit 13 updates the last use date and time added to the subscriber data stored in the cache storage unit 14 to the current time.

  On the other hand, when the designated subscriber data does not exist in the cache storage unit 14 (S13: NO), the data acquisition unit 12 accesses the subscriber database 2 and acquires (downloads) the subscriber data. The subscriber data thus stored is stored in the cache storage unit 14 (S16). The service providing unit 11 performs a service process to be provided to the terminal 4 using the subscriber data acquired by the data acquiring unit 12.

  Then, the priority control unit 13 sets the initial value P0 as the priority P and the current time as the last use date and time for the subscriber data stored in the cache storage unit 14 (S17).

  When the new subscriber data acquired from the subscriber database 2 in S16 is stored in the cache storage unit 14, when the capacity of the cache storage unit 14 is insufficient, that is, a space for storing the subscriber data. When the capacity does not exist in the cache storage unit 14 (when the data size of the subscriber data exceeds the capacity of the free storage area of the cache storage unit 14), the data acquisition unit 12 has a free capacity in which the subscriber data can be stored. Until data can be secured, the data with the lower priority P is deleted in order, and the subscriber data is stored.

  Next, changes in the priority of cache data stored in the cache storage unit 14 will be described.

  FIG. 5 is an image diagram showing a change in priority of subscriber data stored in the cache storage unit 14. The priority control unit 13 monotonously decreases the priority P of each data with time according to a predetermined function (f1 (t) or f2 (t)). However, the priority P is P ≧ 0.

  Here, when the service provision time is known from the subscriber data, it is possible to change the priority reduction rate of each data in each time zone by using different functions within the service provision time and outside the service provision time. Good. For example, in the billing service, the subscriber data has a time zone in which the terminal can be connected to the destination terminal, and the connectable time zone is determined to be within the service provision time and the connection unavailable time zone is determined to be outside the service provision time It is possible to do.

  In the example shown in FIG. 5, the priority is decreased with the passage of time according to the function of f1 (t) within the service providing time. In addition, outside the service providing time, the priority P is decreased with the passage of time according to a function of f2 (t) different from f1 (t). In FIG. 5, the functions f1 (t) and f2 (t) are set so that the speed of lowering the priority outside the service provision time is faster than the speed of lowering the priority within the service provision time. .

  In the case of a service whose service provision time is unknown from the subscriber data, only the function f1 (t) is always used. Also, f1 (t) and f2 (t) may be the same function.

  In FIG. 5, the time period from t4 to t6 is outside the service provision time, and the other time periods from t1 to t4 and t6 to t7 are within the service provision time. First, new subscriber data acquired from the subscriber database 2 at time t1 is stored in the cache storage unit 14 (FIG. 4: S16). At this time, the initial value P0 is set as the priority of the data (FIG. 4: S17). Then, between t1 and t2, the priority is decreased using the function f1 (t) within the service provision time. Then, the data stored in the cache storage unit 14 at time t2 is used (FIG. 4: S14), and the priority P is increased by adding the increase value Q to the priority P of the data at time t2. (FIG. 4: S15).

  Then, between t2 and t3, the priority is decreased using the function f1 (t), and when the data is used at time t3, the increase value Q is added to the priority P of the data at time t3. Thus, the priority P is increased, and the priority is decreased between t3 and t4 using the function f1 (t).

  Since the period from t4 to t6 is outside the service providing time, the priority is reduced after t4 using the function f2 (t) outside the service providing time. The data is used at the time t5, and the priority value P is increased by adding the increment value Q to the priority P of the data at the time t5, and the priority is obtained using the function f2 (t) between t5 and t6. Decrease.

  Since the period between t6 and t7 is within the service providing time, the priority is reduced using the function f1 (t) after t6. At time t7, since the priority P is small, it is assumed that the data is deleted from the cache storage unit 14 in order to store other new data in the cache storage unit 14.

  In this embodiment, when the original subscriber data in the subscriber database 2 is updated, an update notification is transmitted from the subscriber database 2 to the application server 1, and the data acquisition unit 12 of the application server 1 When the updated subscriber data exists in the cache storage unit 14, the subscriber data is updated. Note that the priority value and the last use date are not changed during the update.

<Example of payment service>
Next, as an example of the present embodiment, a case in which subscriber data for a billing service is created in the cache storage unit 14 will be described.

  FIG. 6 shows an example of subscriber data stored in the subscriber database 2 of the present embodiment. The subscriber database 2 shown in the figure has a subscriber table 21 and a service time table 22 provided for each subscriber as subscriber data. The subscriber table 21 stores a subscriber ID, a logical number, a physical number, and a service providing time in association with each subscriber. Here, it is assumed that a pointer to the service time table 22 of the subscriber ID is set in the service provision time. The service time table 22 may be integrated into the subscriber table 21 by setting the data of the service time table 22 to the service provision time of each subscriber ID in the subscriber table 21.

  Next, the operation of the receipt / charge service of this embodiment will be described.

  The application server 1 converts the telephone number (logical number) set in the Request-URI of the INVITE signal received from the call control server 3 into the telephone number (physical number) of 0AB-J. Return to The call control server 3 transmits an INVITE signal to the destination terminal 4 using the converted telephone number 0AB-J. Further, it is assumed that the service provision time is set in the subscriber data for the billing service as shown in FIG. If the application server 1 receives an INVITE signal from the call control server 3 when it is outside the service provision time, it sends a guidance to the calling terminal 4 that it is “out of service provision time”. The guidance may be sent to a device other than the application server 1 (for example, a media server for guidance reproduction).

  FIG. 7 and FIG. 8 are sequence diagrams showing an example of the operation of the receipt / charge service in this embodiment. FIG. 7 is a sequence showing an operation example within the service provision time of the receipt / charge service. Note that the two call control servers 3 shown in FIG. 7 are the same server.

  The calling terminal 4 sends the terminal 4 that is the target of the billing service as the destination. As a result, an INVITE signal in which 0120 indicating that the service is a charged service is set in the Request-URI (destination) is transmitted to the call control server 3 (S21). Since 0120 is set in the Request-URI, the call control server 3 transmits the INVITE signal to the application server 1 (S22).

  The application server 1 converts the telephone number (logical number) starting from 0120 set in the Request-URI into a telephone number (physical number) of 0AB-J and returns it to the call control server 3 (S23). That is, the service providing unit 11 of the application server 1 requests the data acquisition unit 12 to acquire subscriber data having the logical number. The data acquisition unit 12 acquires the instructed subscriber data by the process described with reference to FIG. Here, it is assumed that subscriber data (including service provision time) whose logical number is “0120XXXXXX” shown in FIG. 6 is acquired from the subscriber database 2 or the cache storage unit 14. The service providing unit 11 compares the service providing time included in the subscriber data acquired by the data acquiring unit 12 with the current time. Here, since it is within the service provision time, the logical number set in the Request-URI of the INVITE signal is converted into the physical number of the acquired subscriber data and returned to the call control server 3.

  The call control server 3 transmits an INVITE signal to the called terminal 4 having the converted physical number (S25). When the destination terminal 4 responds, a 200: OK message is transmitted to each of the call control server 3, the application server 1, and the calling terminal 4 (S26), a session is established between the calling terminal 4 and the called terminal 4, and a call starts. (S27).

  FIG. 8 is a sequence showing an example of an operation outside the service provision time of the receipt / charge service. Similar to S21 and S22 of FIG. 7, the INVITE signal is transmitted to the application server 1 (S31, S32). The application server 1 acquires the subscriber data of the logical number set in the Request-URI, compares the service provision time of the acquired subscriber data with the current time (S33), and here, outside the service provision time Therefore, without performing the number conversion process, a 200: OK message is transmitted to establish a session with the calling terminal 4 (S34, S35), and a guidance notifying that it is out of service time is sent to the calling terminal. (S36).

  In FIG. 7 and FIG. 8, the 1xx response is omitted.

  FIG. 9 is an image diagram for explaining a change in priority of the subscriber ID 1234567 having the telephone number “0120xxxxxx” shown in FIG.

  In FIG. 9, it is assumed that the service provision time is until 18:00 and that the service provision time is not after 18:00. Moreover, in FIG. 9, the priority control part 13 shall control the priority P using the following values and functions.

・ Initial value: P0 = 100
・ Function within service provision time: f1 (t) = t
-Function outside service provision time: f2 (t) = 2t
Increased value Q function: g (s, t) = 1/2 {P0− (t−s)}
However, if P0 <(t−s), g (s, t) = 0
Note that t and s are variables that express time in minutes. For example, 17:30 becomes 1050, and 18:10 becomes 1090.

  First, at 17:10, in the state where the subscriber data of “subscriber ID 1234567” is not cached in the cache storage unit 14 of the application server 1, the application server 1 sets an INVITE signal in which “0120xxxxxx” is set in the Request-URI. Is received (S41). Since there is no subscriber data in the cache storage unit 14, the application server 1 downloads the subscriber data of the subscriber ID “1234567” from the subscriber database 2, stores it in the cache storage unit 14, and performs number conversion. Do. At this time, an initial value P0 (= 100) is set as the priority P.

  Thereafter, at 17:30, the application server 1 receives again the INVITE signal in which 0120xxxxxx is set in the Request-URI (S42). Since the cache storage unit 14 has the subscriber data, the application server 1 performs number conversion processing using the subscriber data. At this time, the priority P is subtracted according to the function f1 (t) and decreased to 80 (that is, 20 minutes have passed from 17:10 to 17:30, so 100-20 = 80) Become). As a result of the access to the subscriber data, the calculation result (40) of the function g (17:10, 17:30) is added to the current priority P (80). It becomes.

  Thereafter, at 17:40, the application server 1 receives the INVITE signal in which 0120xxxxxx is set in the Request-URI again (S43), and performs number conversion processing using the subscriber data in the cache storage unit 14. At this time, the priority P is subtracted according to the function f1 (t) and decreased to 110. As a result of the access to the subscriber data, the calculation result (45) of the function g (17:30, 17:40) is added to the current priority P (110). 155.

  Since it is outside the service provision time after 18:00, the priority P is subtracted according to f2 (t).

  Then, at 18:30, the application server 1 receives again the INVITE signal in which 0120xxxxxx is set in the Request-URI (S44), and here, because it is out of service time, sends out guidance. At this time, the priority P is subtracted according to the function f2 (t) after 18:00 and decreased to 75. As a result of the access to the subscriber data, the calculation result (25) of the function g (17:40, 18:30) is added to the current priority P (75) to be 100.

  In the present embodiment described above, the priority of each data stored in the cache storage unit 14 is lowered as time passes, and when access to the data occurs, the priority of the data is increased. As a result, in the present embodiment, even for data whose access frequency increases during a certain period and thereafter decreases, the cache storage unit 14 appropriately deletes the data from the cache according to the usage status of the data. This area can be used effectively.

  In this embodiment, when access to data in the cache storage unit 14 occurs, an increase value is calculated using the last access date and time and the current time of the data, and the priority is set by adding the increase value. increase. Thus, by calculating the value to be added to the priority using the time from the last access date to the next access, the priority can be appropriately controlled according to the data usage status.

  Further, in this embodiment, when the data service provision time can be determined, the speed at which the priority is lowered outside the service provision time is set faster than the speed at which the priority is lowered within the service provision time. In this way, for data outside the service provision time, by increasing the speed of decreasing the priority, it is possible to prevent data with low usage frequency outside the service provision time from remaining in the cache storage unit 14 without being deleted. Can do.

  In addition, this invention is not limited to the said embodiment, A various change and application are possible within a claim.

DESCRIPTION OF SYMBOLS 1: Application server 11: Service provision part 12: Data acquisition part 13: Priority control part 14: Cache memory | storage part 2: Subscriber database 3: Call control server 4: Terminal

Claims (6)

An application server,
Cache storage means for storing data;
Priority control means for controlling the priority of each data stored in the cache storage means;
A deletion unit that deletes data in descending order of priority when the capacity of the cache storage unit is insufficient,
The priority control means decreases the priority of data stored in the cache storage means as time elapses, and when access to data occurs, increases the priority of the data,
The priority control means makes the speed of lowering the priority with the passage of time outside the service provision time faster than the speed of lowering the priority with the passage of time within the service provision time. . The application server according to claim 1,
When the access to the data occurs, the priority control means calculates an increase value using the last access date and time and the current time of the data, and adds the increase value to increase the priority. Application server. The application server according to claim 1 or 2 ,
The data stored in the cache storage means is subscriber data,
When the subscriber data to be used in the request received from the call control server does not exist in the cache storage means, the apparatus further comprises data acquisition means for reading the subscriber data from the subscriber database and storing it in the cache storage means. An application server characterized by A cache control method performed by an application server including a cache storage unit,
A priority control step for controlling the priority of each data stored in the cache storage unit;
When the capacity of the cache storage unit is insufficient, a deletion step of deleting data in order from low priority data is performed,
The priority control step decreases the priority of data stored in the cache storage unit as time elapses, and when access to the data occurs, increases the priority of the data ,
The priority control step is configured to make a speed of lowering the priority as time elapses outside the service provision time faster than a speed of lowering the priority as time elapses within the service provision time. Method. The cache control method according to claim 4 ,
In the priority control step, when access to data occurs, an increase value is calculated using the last access date and time and the current time of the data, and the priority is increased by adding the increase value. Cache control method. A cache control method according to claim 4 or 5 ,
The data stored in the cache storage unit is subscriber data,
When the subscriber data used in the request received from the call control server does not exist in the cache storage unit, the subscriber data is further read out from the subscriber database and stored in the cache storage unit. A cache control method characterized by the above.

Images