JPH08166916A - Online data delivery system - Google Patents

Abstract

PURPOSE: To easily optimize data transfer quantity to a low-order server and to easily correspond to the change of environment on a delivery route by changing the identifier set of partial data stored in a data base in accordance with the data identifier set received from the low-order server. CONSTITUTION: When the low-order servers 103 and 105 input the identifiers of partial data, which a user desires, the low-order servers 103 and 105 periodically collect the identifiers for the whole users, which are inputted by the low- order servers, derive the partial data identifier set as the sum set and transfer the partial data identifier set to high-order servers 102 and 106. On the other hand, the high-order servers 102 and 106 change the identifier set of partial data stored in the data base in accordance with the partial data identifier set received from the low-order servers 103 and 105. Thus, manual work for designating data to be delivered is eliminated.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an online data delivery system, and in particular, it continuously transfers data from an upper server to a lower server via a network and sequentially stores the transferred data on the lower server side. The present invention relates to an online data delivery system in which the user can access the data that the user needs by accessing the stored data.

[0002]

2. Description of the Related Art Conventionally, data created by many users and computer systems are sequentially collected via a network, and the collected data is copied and delivered to a plurality of server systems widely distributed on the same network. There is an online data delivery system that makes it possible for many users to refer to data of the same content in a short time in response to an access request from users on a server.

Conventionally, in such a system, in order to execute delivery at a sufficient speed for a huge number of users, each server sets a transfer route having a graph structure, and one of the servers creates and delivers it. It has a mechanism to deliver the started data to other servers according to the transfer route of this graph structure. In this case, the transfer path of the graph structure is an omnidirectional transfer path having no parent-child relationship.

Further, all the data are classified into a plurality of partial data according to their contents, an identifier is attached to each partial data when the partial data is created, and one server (upper server) is adjacent to another server (lower server). By limiting the partial data to be delivered to the server by its identifier, a mechanism for suppressing the amount of data transfer between servers is provided.

[0005]

However, in the conventional online data delivery system as described above,
(1) How to specify the partial data to be transferred to the adjacent server,
(2) There are the following problems in terms of optimizing the transfer amount of partial data.

That is, in the conventional system, designation of partial data to be transferred from one server to another adjacent server is performed by registering the identifier of each partial data in a specific database, and at the time of transfer between servers, Refer to the database and classify it into one that transfers all data and one that does not transfer, and the transfer process is not executed for those that do not.

However, since the contents of the database are updated manually by the administrator, even if there is a change in the environment on the delivery route ahead of the lower server, for example, there is a change in the contents of data desired to be referenced, that change. Is difficult to reflect in the specified contents of the transfer part data.

Further, the selection of the partial data to be transferred is conventionally made by the judgment of the administrator, and there is a possibility that unnecessary data may be transferred or necessary data may be lost due to an error in the judgment of the administrator. In order to avoid this, the administrator must always keep the partial data to be delivered from the own server the same as the partial data required by the lower server.

However, the partial data required by the subordinate server refers to the partial data required by all the servers on all routes to which the subordinate data is delivered and the subordinate server is accessed to refer to the partial data. Since it is a union with the reference partial data setting contents of the user, it becomes very difficult to judge the partial data to be delivered by manual processing. That is, there is a problem that it is very difficult to optimize the data transfer amount to the lower server.

An object of the present invention is to provide an online data delivery system which can easily optimize the amount of data transferred to a lower server and can easily cope with changes in the environment on the delivery route. It is in.

[0011]

In order to achieve the above object, the present invention provides a method in which data is classified into a plurality of partial data according to the contents thereof as information for designating data to be transferred from an upper server to a lower server. Database that stores a set of identifiers of the above, input means for inputting the identifier of the partial data that the user who uses the lower level server wants to refer to in the lower level server, and the input identifiers on the lower level server on a regular basis. Means for collecting the partial data identifier set as a union and transferring the partial data identifier set to the upper server, and storing in the database according to the partial data identifier set received from the lower server in the upper server And a means for changing the identifier set of the generated partial data.

Further, in response to a reference request for partial data not stored in the lower server, an irregular delivery request including the identifier of the partial data is transmitted to the upper server, and the corresponding partial data is irregularly sent from the upper server. It is characterized by comprising means for delivering to.

Further, the frequency of the irregular delivery request is monitored in the upper server, and when the partial data amount of the irregular delivery exceeds a predetermined value, the partial data to be continuously delivered,
It is characterized by comprising means for registering the identifier in the database.

When data is delivered by a plurality of servers having a graph structure, a set of identifiers of partial data that must be transferred from one server to another adjacent server other than the server that transfers the partial data. It is characterized in that it is provided with means for notifying it to the adjacent upper-layer server and repeating this operation by sequentially tracing the graph structure.

[0015]

According to the present invention, (1) When the user inputs the identifier of the partial data desired by the user in the lower server, the lower server periodically uses the identifiers of all the users input in the lower server on the lower server. Data, the partial data identifier set as a union thereof is derived, and the partial data identifier set is transferred to the upper server.

On the other hand, the upper server changes the identifier set of the partial data stored in the database according to the partial data identifier set received from the lower server.

This eliminates the need for manual work to specify the data to be delivered. Further, the amount of data transferred from the upper server to the lower server can be optimized to the minimum within the range that the user desires to refer to.

(2) In response to a reference request for partial data not stored in the lower server, the lower server sends an irregular delivery request including the identifier of the partial data to the upper server.

As a result, it is not necessary to constantly deliver from the upper server to the lower server, but for the partial data that the user desires to refer to irregularly, the upper server delivers the partial data to the lower server at any time. You can

(3) The frequency of the irregular delivery request is monitored in the upper server, and if the partial data amount of the irregular delivery exceeds a predetermined value, the identifier is registered in the database as the partial data to be continuously delivered.

As a result, when the frequency of irregular delivery requests for specific partial data is high, the corresponding partial data will be constantly delivered and held on the lower server. The user does not need to transfer the data one by one when referring to it, and the waiting time of the user is shortened and the operability is improved.

(4) When partial data is delivered by a plurality of servers having a graph structure, an identifier set of partial data that must be transferred from one server to another adjacent server is transferred to the partial data. The operation of notifying adjacent upper servers other than the server is repeated along the route of the graph structure.

This makes it possible to minimize the total amount of data delivered between a plurality of servers forming a graph structure. Therefore, the total amount of data exchanged in the entire network is reduced, and the storage device on the server and the line cost are saved.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below with reference to illustrated embodiments.

FIG. 1 is an overall configuration diagram showing an embodiment of an online data delivery system according to the present invention.

In the figure, 101 to 106 are online data delivery servers which are connected to the networks 120 to 125 and which transmit and receive data between the servers via the networks.

107 to 112 are delivery servers 101 to 1
It is an auxiliary storage device for storing data to be transmitted and received by 06 and a database for storing information related thereto.

Reference numeral 113 is a data creation unit for creating / transmitting data delivered by the delivery system. In this embodiment, it is connected to the delivery server 101, but may be connected to any of the other delivery servers 102 to 106.

Reference numerals 114 and 115 denote the delivery server 10.
A user terminal device that takes out data from 3, 104, performs an output operation such as display on a screen or printing on a paper, or receives an operation such as a data reference request from a user, and a portion desired to be referred. When accessing data, the identifier of the partial data is input using the user terminal device.

In this embodiment, only the delivery servers 103 and 104 are connected, but all other delivery servers may be connected.

FIG. 2 is a data structure diagram showing the structure of data transmitted and received between delivery servers. Delivery data is classified according to its contents, and a partial data identifier 201 is an identifier of partial data obtained by the classification. And the partial data 202 are repeated. The partial data identifier 201 indicates the classification of the contents of the partial data 202, and the partial data 202 indicates a group of data having meaning independently.

FIG. 3 shows the contents of the delivery partial data registration database 300.
One is stored in each of the delivery servers 101 to 106 and 0 in each of the auxiliary storage devices 107 to 112.

The database 300 stores a set of identifiers of partial data as information designating data to be transferred from the upper server to the lower server.

In this database 300, 301
Reference numerals a to 301c are identifiers for identifying a delivery server (upper server) holding the database 300 and another adjacent delivery server (lower server) connected via a network.

Reference numeral 302 is an identifier list showing the identifiers 201 of the partial data to be regularly delivered to the delivery servers (lower servers) corresponding to the identifiers 301a to 301c. In the example of FIG. 1) For the adjacent delivery server 301a, partial data identifiers (1)-
It is registered that the partial data of (7) should be delivered, and the partial data of partial data identifiers (1), (3), and (7) are delivered to the adjacent delivery server of the server identifier (2) 301b. It should be registered that the partial data of partial data identifiers (1), (2), (3) and (7) should be delivered to the adjacent delivery server of server identifier (3) 301c. ing.

FIG. 4 shows the structure of all the partial data received by the delivery servers 101 to 106 and stored in the auxiliary storage devices 107 to 112.
401c is a partial data identifier corresponding to the partial data identifier 201 in FIG.

Reference numeral 402 is a serial number assigned to each partial data in the partial data group classified by a specific identifier.

Reference numeral 403 is each partial data classified by a specific identifier.

FIG. 5 shows a user identifier list 500, which uses the user terminal devices 114 and 115 connected to the delivery servers 103 and 104 to access the delivery servers 101 to 106. , Each delivery server 10
5 is a list of identifiers 501 of all users who take out and refer to the partial data stored in the auxiliary storage devices 107 to 112 of Nos. 1 to 106 in the configuration of FIG.

FIG. 6 shows an identifier list 600 of partial data referred to by a user. This is an auxiliary for each of the delivery servers 101 to 106 corresponding to all the users who access it. Storage device 107
˜112, the identifiers 602 of the partial data that each user wants to refer to are arranged for each user identifier 601.

FIG. 7 shows the structure of a history database 700 of irregular delivery requests for partial data.

This is a record of the contents of irregular delivery requests for partial data transmitted from the lower servers in the past in each of the delivery servers 101 to 107, and the lower server identifiers 701a, 701b, 701c, ... The delivery request history information including the delivery request reception date and time 702 and the delivery requested partial data identifier 703 is recorded.

8 to 10 are flow charts showing the flow of processing of this embodiment. The operation of this embodiment will be described below with reference to this flow chart.

First, the steady data delivery process will be described with reference to the flowchart of FIG.

The data delivered by the system of this embodiment is created by the data creating unit 113. The created data is classified into semantically independent partial data 202. Prior to delivery to the delivery server 101, an identifier 201 according to the classification of the content is added to each of the partial data 202, and the whole data is obtained. As a result, data having the structure shown in FIG. 2 is created.

The data created by the data creating unit 113 is received by the delivery server 101 (step 801).

The delivery servers 102 and 106, which are lower servers as viewed from the delivery server 101, receive the data of the structure of FIG. 2 received in step 801 from the delivery server 101 which is the upper server (step 802).

The received partial data 202 is recorded in the auxiliary storage devices 108 and 112 in the format shown in FIG. 4 with reference to the partial data identifier 201 (step 803).

Next, the delivery servers 102 and 106
Referring to the delivery partial data registration database 300 of FIG. 3, the identifier 301a of the subordinate server (subordinate server of the delivery servers 102 and 106) to which each of the partial data 202 stored in the auxiliary storage devices 108 and 112 in step 802 should be delivered. ˜301c are checked (step 804).

Next, when the identifier of the lower server (103 for 102 and 103 and 105 for 106) to be delivered is found, the identifier of the partial data to be delivered to the lower server of that identifier is determined. Extracted from the partial data identifier list 302 of FIG. 3, the partial data of the extracted identifier is extracted from all the partial data stored in the auxiliary storage devices 108 and 112 in the format of FIG. 4, and delivered to the corresponding lower server. (Step 806).

Next, the process of step 806 is repeated until the partial data has been delivered to all of the lower servers checked in step 804 (step 805).

The processing from step 802 to step 806 is performed until all the partial data created by the data creating unit 113 is delivered to all delivery servers that must be delivered according to the delivery partial data registration database 300 of FIG. It is recursively repeated from the server to the lower servers.

Next, the delivery partial data registration database 3
The update process of 00 will be described with reference to the flowchart of FIG.

This process is performed by regarding a specific server as an upper server and a plurality of servers connected thereto as lower servers.

First, the upper server side refers to the delivery partial data registration database 300 to identify the lower server to which the partial data 202 should be delivered from the server as the server identifier 30.
List all from 1a to 301c (step 9)
01).

The following processing is executed for each of these lower servers.

Next, a delivery database update request is transmitted to one of the lower servers (step 90).
3).

The subordinate server that has received this delivery database update request uses the list of user identifiers 501 in FIG. 5 to display the reference partial data list 600 for each user for all users who are accessing the same. The partial data identifier list 30 is referred to by referring to the delivery partial data registration database 300 held by the user.
2 and the partial data identifiers listed in the partial data identifier 602 are created (step 904).

The list of partial data identifiers created here is transmitted to the host server which has transmitted the delivery database update request (step 905).

The upper server refers to the partial data identifier list received from the lower server and reflects it in the identifier column of the corresponding lower server in its own delivery partial data registration database 300 (step 906).

That is, the partial data identifiers listed in the partial data identifier list received from the subordinate server but not recorded in the delivery partial data registration database 300 are additionally registered in the identifier list 302, and the distribution is performed. The partial data identifiers registered in the partial data registration database 300 but not listed in the identifier list 302 are deleted.

The above processing is executed for all lower servers connected to the corresponding upper server (step 90).
2), the process ends.

This process may be executed independently of the data delivery process and independently among the servers 101 to 106. For example, it may be executed as a batch process so as to be executed periodically, or an administrator may The terminal device 114 by hand,
Alternatively, the command may be input from 115.

Next, with reference to the flowchart of FIG. 10, a process in which the user requests the delivery of the partial data which is not regularly delivered by the above two processes irregularly will be described.

In this process, when the user tries to refer to a specific partial data group from the terminal device 114 or 115, the corresponding partial data group is detected as the terminal device 114, 115.
Is executed when it does not exist in the auxiliary storage device (FIG. 4) of the server (in this embodiment, the delivery servers 103 and 104 in this embodiment) to which is connected.

First, the server 103 or 104 connected to the terminal device 114 or 115 receives a user's partial data reference request from the terminal device 114 or 115 (step 1001).

In the server 103 or 104, the partial data held by itself is changed to the partial data identifier 4 in FIG.
01a to 401c are searched (step 1002), and if partial data corresponding to the requested partial data identifier exists, the partial data (there may be a plurality) is transmitted to the terminal device 114 or 115. (Step 1004).

In step 1002, if the requested partial data does not exist on the server 103 or 104, the partial data delivery request is sent using the corresponding partial data identifier.
05 to the upper server (step 100)
3).

The upper server receives the partial data delivery request from the lower server 104 or 105 (step 1005).

The upper-level server refers to the history database 700 of FIG. 7 and obtains the total number of requests for partial data of the same identifier received from the corresponding server in the past within a certain fixed period (step 1006).

When this number, that is, the frequency of requests exceeds a certain value (step 1007), the partial data identifier 302 is registered in the corresponding lower server identifier column of the delivery partial data registration database 300 of the server ( Step 1008).

Here, the fixed value is set by the administrator, and the reception frequency or the reception amount per unit time of the partial data which the server always receives from the upper server is automatically measured. However, a method of calculating and using a boundary value from which transfer cannot be performed at a speed sufficient for the line capacity can be considered.

Next, it is checked by searching all the partial data in the format of FIG. 4 whether or not the corresponding partial data exists in the auxiliary storage device. If it exists, the corresponding partial data is taken out and delivered to the lower server that issued the request (step 1011).

If the partial data does not exist, the partial data delivery request is transmitted to the upper server for the server using the partial data identifier attached to the partial data (step 1010).

In response to this request, the partial data delivered from the host server is delivered to the terminal device 114 or 115 of the user while tracing the transmission of the request in the reverse order.

[0076]

As described above, according to the present invention,
Data created by many users and computer systems are sequentially collected over the network, and this data is copied and delivered to multiple server systems widely distributed on the network, and multiple users access each server. By doing so, in the online data delivery system in which the same content data is distributed to a large number of users in a short time and referred to, (1) The user wants to refer to the total amount of data delivered from the upper server to the lower server. Can be minimized in range.

(2) It is not necessary to constantly deliver from the upper server to the lower server, but partial data that the user desires to refer to irregularly is delivered from the upper server to the lower server at any time. You can

(3) When the frequency of delivery requests for specific partial data is high, the corresponding partial data will be constantly delivered and held on the lower server, so that the user can send the corresponding partial data. There is no need to transfer the data one by one when referring to it, and the waiting time for the user is shortened and the operability is improved.

(4) It is possible to minimize the total amount of data delivered between a plurality of servers forming a graph structure.
Therefore, the total amount of data exchanged over the entire network is reduced, and there is an effect that the storage device on the server and the line cost can be saved.

[Brief description of drawings]

FIG. 1 is a system configuration diagram showing an embodiment of the present invention.

FIG. 2 is a structural diagram of delivery data.

FIG. 3 is a configuration diagram of a registration database of delivery partial data.

FIG. 4 is a configuration diagram of all delivery data stored in an auxiliary storage device.

FIG. 5 is a diagram showing a user identifier list for each server.

FIG. 6 is a diagram showing a reference partial data list for each user.

FIG. 7 is a data configuration diagram of a history database of irregular data delivery requests.

FIG. 8 is a flowchart showing a procedure of a steady data delivery process.

FIG. 9 is a flowchart showing a process of updating a partial data registration database.

FIG. 10 is a flowchart showing a process for an irregular delivery request.

[Explanation of symbols]

101 to 107 ... Delivery server, 107 to 112 ... Auxiliary storage device, 113 ... Data creation unit, 114, 115 ... User terminal device, 201 ... Partial data identifier, 202 ... Partial data, 300 ... Delivery partial data registration database, 3
01a-301c ... Server identifier.

Continuation of front page (51) Int.Cl. ⁶ Identification code Office reference number FI technical display location H04L 12/58

Claims (4)

[Claims] 1. A plurality of uses by delivering data from an upper server to a lower server via a network, sequentially accumulating data received by the lower server, and accessing the data accumulated on the lower server. In an online data delivery system that allows users to refer to the data within the range requested by each person, as information that specifies the data that should be transferred from the upper server to the lower server, when the data is classified into multiple partial data A database that stores a set of identifiers, an input unit that inputs the identifier of the partial data that the user who uses the lower-level server wants to reference in the lower-level server, and the input identifiers on the lower-level server on a regular basis. , And derive a partial data identifier set as the union of the Online data delivery, comprising: means for transferring the partial data to the upper server, and means for changing the identifier set of the partial data stored in the database according to the partial data identifier set received from the lower server in the upper server. system. 2. In response to a reference request for partial data not stored in the lower server, an irregular delivery request including the identifier of the partial data is transmitted to the upper server, and the corresponding partial data is irregularly sent from the upper server. The online data delivery system according to claim 1, further comprising means for delivering to the online data delivery system. 3. The upper server monitors the frequency of the irregular delivery request transmitted from the lower server, and if the partial data amount of the irregular delivery exceeds a predetermined value, the identifier is used as the partial data to be continuously delivered. The online data delivery system according to claim 2, further comprising means for registering in the database. 4. When partial data is delivered from a higher-level server that is a source of partial data by a route having a graph structure using a plurality of transfer servers, it must be delivered from one server to another adjacent server. 2. A means for notifying a set of identifiers of partial data that does not become a part of the adjacent server other than the server that delivers the partial data and repeating this operation by sequentially tracing the graph structure. 3) An online data delivery system according to any one of 3).