JP2018036952A - Distribution device and distribution method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a distribution DB system with high performance by reducing communication between DBs.SOLUTION: There is provided a distribution DB system in which an acquisition part 15a stores pieces of data belonging to a same community in a same DB, and distributes pieces of data belonging to different communities into plural DBs and stores the same and acquires a communication traffic volume following to request of data to the DB from each community, a determination part 15c determines a combination of communities stored in the same DB, in an order in which the communication traffic volume is large among communities. In addition, a creation part 15b creates a community sub group by combining the plural communities, reduces variation of the communication traffic volume from each community sub group in all community sub groups, and the determination part 15cc may determine a combination of the communities stored in the same DB in an order in which the communication traffic volume is large among the created community sub groups.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a dispersion apparatus and a dispersion method.

  In general, a distributed DB system is known in which large-scale data is distributed and accommodated in a plurality of databases (DB). For example, in a system in which a predetermined Web server executes a process for a user request for each request, data necessary for processing each request is accommodated in the DB of each Web server (see Patent Document 1).

  In a distributed DB system, there is a case where data necessary for processing a request is not accommodated in a DB of a server from which a user has requested processing. In that case, it is necessary to acquire the desired data by performing inter-DB communication in cooperation between the server requested by the user and the server in which the desired data is accommodated. For example, in the telephone service, when subscriber data is accommodated in a DB prepared for each prefecture, the data of the destination user is acquired by inter-DB communication during a call between prefectures.

JP 2014-096113 A

  However, when the amount of communication between DBs increases and congestion occurs, a delay from a request to the output of a processing result, that is, a TAT (Turn Around Time) delay or a processing failure may occur.

  The present invention has been made in view of the above, and an object of the present invention is to provide a highly efficient distributed DB system by reducing inter-DB communication.

  In order to solve the above-described problems and achieve the object, the distribution device according to the present invention accommodates data belonging to the same community in the same database and distributes data belonging to different communities to a plurality of databases. In the database system to be accommodated, an acquisition unit that acquires a communication amount associated with a data request from each community to the database, and the acquired communication amount, the communication amount between the communities is stored in the same database in descending order. And a determining unit that determines a combination of communities to perform.

  According to the present invention, it is possible to provide a highly efficient distributed DB system by reducing inter-DB communication.

FIG. 1 is an explanatory diagram for explaining a processing target of a distribution apparatus according to an embodiment of the present invention. FIG. 2 is an explanatory diagram for explaining an outline of processing of the distribution apparatus according to the present embodiment. FIG. 3 is a schematic diagram illustrating a schematic configuration of the dispersion apparatus according to the present embodiment. FIG. 4 is an explanatory diagram for explaining the processing of the creation unit. FIG. 5 is an explanatory diagram for explaining processing of the creation unit. FIG. 6 is an explanatory diagram for explaining the processing of the determination unit. FIG. 7 is an explanatory diagram for explaining the processing of the determination unit. FIG. 8 is an explanatory diagram for explaining the effect of the distributed processing. FIG. 9 is a flowchart showing the distributed processing procedure of this embodiment. FIG. 10 is a diagram illustrating a computer that executes a distributed program.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. In addition, this invention is not limited by this embodiment. Moreover, in description of drawing, the same code | symbol is attached | subjected and shown to the same part.

[Distributed device processing overview]
First, a processing target of the distribution apparatus according to this embodiment will be described with reference to FIG. In this embodiment, for example, the distributed device targets a distributed DB system that accommodates telephone service subscriber data distributed in a plurality of DBs in units of communities such as prefectures to which the subscriber belongs. In the distributed DB system 1 illustrated in FIG. 1, data belonging to community A, community B, and community C is accommodated in DB # 1, and data belonging to community D, community E, and community F is accommodated in DB # 2. .

  Here, when the subscriber belonging to the community A makes a call to the subscriber belonging to the community B, the data of the destination subscriber acquired is accommodated in the same DB # 1 as the data A belonging to the community A. Yes. Thus, when the data B belonging to the community B is requested from the community A, the desired data B is accommodated in the same DB as the DB # 1 requested by the requesting community A. As shown by the broken line, it is acquired by intra-DB communication in DB # 1.

  On the other hand, when data belonging to the community D is requested from the community C, such as when a subscriber belonging to the community C makes a call with a subscriber belonging to the community D as a destination, the desired data D is the DB requested by the community C. It is stored in DB # 2, which is different from # 1. Therefore, the desired data D is acquired by performing inter-DB communication in cooperation between DB # 1 and DB # 2, as indicated by a one-dot chain line in FIG.

  While such inter-DB communication is performed, the request source process is awaited. Further, when the communication amount between DBs increases and congestion occurs, there is a possibility that a TAT delay occurs or the process ends in failure. Therefore, the distributed apparatus according to the present embodiment determines the combination of communities that store data in the same DB so that the data of communities having a large amount of communication between communities is stored in the same DB. Reduce intercommunication. The community is exemplified by a city, a company, a department, etc. in addition to a prefecture.

  Next, with reference to FIG. 2, an outline of processing of the distribution apparatus according to the present embodiment will be described. FIG. 2A shows the amount of communication associated with a request for data within the community of each community and the amount of communication associated with the request for data between communities. In FIG. 2A, the requesting communities are shown in the column direction, and the requesting communities are shown in the row direction. FIG. 2A illustrates, for example, that the intra-community traffic volume of the community A is 1 and the inter-community traffic volume from the community A to the community C is 8.

  Here, intra-DB communication includes communication within a community and communication between communities accommodated in the same DB. FIG. 2B illustrates the amount of communication that can be the intra-DB communication of each community out of the amount of communication illustrated in FIG. For example, for the community A, the amount of communication that can be in-DB communication is totaled as 1 + 1 + 8 + 7 + 7 + 2 = 26, where the request source in FIG.

  The inter-DB communication includes communication between communities housed in different DBs. FIG. 2C illustrates the amount of communication between the communities among the amount of communication illustrated in FIG. That is, FIG. 2C illustrates the amount of communication that can be communication between DBs. For example, the inter-community traffic at the intersection of the community A and the community C is 8 as described above.

  As long as the distributed DB system 1 operates, intra-DB communication cannot be reduced. However, if the intra-DB communication is biased to one of the DBs, there is a possibility that the intra-DB communication may be congested. Therefore, the distribution apparatus according to the present embodiment first creates community subgroups in which communities are combined so as to reduce the variation in the amount of communication that can be the intra-DB communication illustrated in FIG.

  Next, for the community subgroups in which each traffic is leveled, the distributed device aggregates the traffic between the community subgroups that can be inter-DB communication as in FIG. Combine community subgroups in descending order. Thereby, the data of communities with much communication volume between communities are accommodated in the same DB, and communication between DB is reduced.

[Distributed device configuration]
Next, a schematic configuration of the dispersion apparatus according to the present embodiment will be described with reference to FIG. As illustrated in FIG. 3, the distributed apparatus 10 according to the present embodiment is realized by a general-purpose computer such as a workstation or a personal computer, and includes an input unit 11, an output unit 12, a communication control unit 13, a storage unit 14, and a control unit. 15. The distribution apparatus 10 executes a distribution process described later, and determines a combination of communities that store data in the same DB so that data of communities having a large amount of communication between communities is stored in the same DB. .

  The input unit 11 is realized using an input device such as a keyboard or a mouse, and inputs various instruction information to the control unit 15 in response to an input operation by the operator. The output unit 12 is realized by a display device such as a liquid crystal display, a printing device such as a printer, an information communication device, and the like, and outputs a result of distributed processing described later to the operator.

  The communication control unit 13 is realized by a NIC (Network Interface Card) or the like, and controls communication between an external device such as a management server and the control unit 15 via a telecommunication line such as a LAN (Local Area Network) or the Internet. .

  The storage unit 14 is realized by a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 14 stores in advance a processing program for operating the distribution apparatus 10, data used during the execution of the processing program, or temporarily stored for each processing. The storage unit 14 may be configured to communicate with the control unit 15 via the communication control unit 13.

  As illustrated in FIG. 3, the control unit 15 performs an acquisition unit 15 a, a creation unit 15 b, and a determination unit 15 c by executing a processing program stored in a memory by an arithmetic processing unit such as a CPU (Central Processing Unit). Function.

  In the distributed DB system 1 that stores data belonging to the same community in the same DB and distributes data belonging to different communities to a plurality of DBs, the acquisition unit 15a responds to a request for data from each community to the DB. Get the accompanying traffic.

  Specifically, the acquisition unit 15a receives, for example, the amount of communication associated with a request for data within the community of each community from the management server of the distributed DB system 1 via the input unit 11 or the communication control unit 13 and between the communities. The amount of communication accompanying the data request (see FIG. 2) is acquired.

  The creation unit 15b creates a community subgroup by combining a plurality of communities, and reduces variation in the amount of traffic from each community subgroup among all the community subgroups among the obtained traffic (see FIG. 2). Let

  Specifically, the processing of the creation unit 15b will be described with reference to FIGS. As illustrated in FIG. 4, the creation unit 15b applies a grouping method described below to create a community subgroup by combining a plurality of communities so as to reduce variation in the traffic of each community subgroup. To do.

  FIG. 5 is an explanatory diagram for explaining the grouping method. Here, as illustrated in FIG. 5A, a case where the data size of the target data of the grouping method sorted in ascending order is represented by y = f (x) and the standard deviation is 0.88 will be described. . In the grouping method, as illustrated in FIG. 5B, the centroid of the difference value of the data size is obtained, and the ratio between the number of data of the data whose difference value of the data size is smaller than the centroid and the number of data of the data larger than the centroid. Is calculated. In the example shown in FIG. 5B, the ratio of the number of data pieces of data smaller than the centroid and the number of data pieces of data larger than the centroid is 26.4 to 13.6, that is, 2 to 1.

  In that case, in the grouping method, as illustrated in FIG. 5C, a subgroup is created by combining three data of two pieces of data smaller than the centroid and one piece of data larger than the centroid. Thereby, the data size of each subgroup can be equalized. In the example shown in FIG. 5C, the data size y of each subgroup is approximately a constant with a, and the standard deviation is 0.12. Thus, according to the grouping method, subgroups can be created so as to reduce variation in data size.

  The determining unit 15c determines the combination of communities accommodated in the same DB in the descending order of the traffic volume between the created community subgroups among the acquired traffic volume (see FIG. 2).

  Specifically, the processing of the determination unit 15c will be described with reference to FIGS. As illustrated in FIG. 6, the determination unit 15 c totals the communication amount between community subgroups for the community subgroup created by the creation unit 15 b. In addition, the determination unit 15c determines a combination of two community subgroups of the request source and the request destination as community groups accommodated in the same DB in descending order of the traffic volume between the community subgroups.

  In the example illustrated in FIG. 6, since the communication amount from the community subgroup 1 to the community subgroup 2 is the largest, it is determined that the community subgroup 1 and the community subgroup 2 are accommodated in the same DB. A community group obtained by combining the determined community subgroup 1 and community subgroup 2 is referred to as a community group A.

  In addition, except for community subgroup 1 and community subgroup 2 that have been determined to be combined, the next largest amount of communication is exemplified by the amount of communication from community subgroup 4 to community group 3. Therefore, it is determined that the community subgroup 3 and the community subgroup 4 are accommodated in the same DB. A community group in which the determined community subgroup 3 and community subgroup 4 are combined is referred to as a community group B.

  As illustrated in FIG. 7, the determination unit 15 c targets the determined community group, and in the same manner as in the example illustrated in FIG. 6, the combination of the two community groups is the same in descending order of the amount of communication between the community groups. It is determined as a community group accommodated in the DB. The determination unit 15c repeats this process to create community groups for the number of DBs to be distributed.

  In the example illustrated in FIG. 7, community subgroup 1 and community subgroup 2 are combined to form community group A. Also, community subgroup 3 and community subgroup 4 are combined to form community group B. Then, since the communication amount between the community group A and the community group B determined as the primary group is the largest, it is determined that the community group A and the community group B are accommodated in the same DB. Similarly, it is determined that the community group D and the community group E are accommodated in the same DB.

  Further, the determination unit 15c outputs the determination result to the output unit 12 in an appropriate format. For example, the communities accommodated in each DB are displayed in a listable manner. In addition, the communication volume between DBs of each DB may be displayed.

  Note that the processing of the creation unit 15b is not necessarily performed. In that case, instead of the community subgroups, the determination unit 15c determines the combination of communities accommodated in the same DB in the descending order of the amount of communication between the communities (see FIG. 2C).

  That is, the determination unit 15 determines a combination of two communities as a community group accommodated in the same DB in descending order of communication amount between the communities. Similarly to the above, for the determined community group, the combination of the two community groups is determined as a community group accommodated in the same DB in the descending order of the communication amount between the community groups. The creation unit 15b repeats this process (see FIG. 7), and creates community groups as many as the number of DBs to be distributed. Even in this case, as illustrated in FIG. 8A, the inter-DB communication amount from DB # 1 to DB # 2 and the inter-DB communication amount from DB # 2 to DB # 1 are reduced as compared to the conventional case. Is done.

  In the present embodiment, the processing of the determination unit 15c is performed after reducing the variation in the amount of communication that can be in-DB communication by the processing of the creation unit 15b. Therefore, as illustrated in FIG. The variation in inter-communication traffic is further reduced.

[Distributed processing]
Next, the distribution process of the distribution apparatus 10 will be described with reference to FIG. The flowchart in FIG. 9 is started, for example, at a timing when there is an operation input instructing the start of distributed processing.

  First, the acquisition unit 15a acquires the communication amount associated with the data request within the community of each community and the communication amount associated with the data request between the communities from the management server of the distributed DB system 1 (step S1). .

  Next, the creating unit 15b creates a community subgroup by combining a plurality of communities so as to reduce the variation in the amount of communication in each community subgroup (step S2).

  Next, the determination part 15c determines the combination of the community accommodated in the same DB in order with much communication volume between the created community subgroups (step S3). Further, the determination unit 15c outputs the result in an appropriate format (step S4). Thereby, a series of distributed processing ends.

  As described above, in the distribution apparatus 10 of the present embodiment, the acquisition unit 15a stores data belonging to the same community in the same DB, and distributes data belonging to different communities to a plurality of DBs. In the distributed DB system 1, the amount of communication associated with the data request for the DB from each community is acquired. Moreover, the determination part 15c determines the combination of the community accommodated in the same DB in order with the largest communication volume between communities among the acquired communication volumes.

  Thereby, according to the distributed process of the distributed apparatus 10 of this embodiment, since communication between DB is reduced, generation | occurrence | production of a TAT delay, a process failure, etc. is suppressed. In this way, a highly efficient distributed DB system can be provided, and restrictions on the number of DBs that can be added can be suppressed.

  In addition, the creation unit 15b creates a community subgroup by combining a plurality of communities, and reduces the variation of the traffic from each community subgroup within all the community subgroups. In that case, the determination unit 15c determines a combination of communities accommodated in the same DB in descending order of communication volume between the created community subgroups. As a result, the variation in the communication amount that can be the intra-DB communication is reduced, so that the variation in the inter-DB communication amount is further reduced.

[program]
It is also possible to create a program in which processing executed by the distributed apparatus 10 according to the above embodiment is described in a language that can be executed by a computer. As an embodiment, the distribution apparatus 10 can be implemented by installing a distributed program that executes the above distributed processing as package software or online software on a desired computer. For example, the information processing apparatus can function as the distribution apparatus 10 by causing the information processing apparatus to execute the above distributed program. The information processing apparatus referred to here includes a desktop or notebook personal computer. In addition, the information processing apparatus includes mobile communication terminals such as smartphones, mobile phones and PHS (Personal Handyphone System), and slate terminals such as PDA (Personal Digital Assistants). In addition, a terminal device used by a user can be a client, and the client can be implemented as a server device that provides the client with services related to the distributed processing. For example, the distributed device 10 is implemented as a server device that provides a distributed processing service that receives a communication amount requesting data from each community and outputs a combination of communities accommodated in the same DB. In this case, the distributed apparatus 10 may be implemented as a Web server, or may be implemented as a cloud that provides services related to the above distributed processing by outsourcing. Hereinafter, an example of a computer that executes a distributed program that realizes the same function as the distributed apparatus 10 will be described.

  As shown in FIG. 10, a computer 1000 that executes a distributed program includes, for example, a memory 1010, a CPU 1020, a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface. 1070. These units are connected by a bus 1080.

  The memory 1010 includes a ROM (Read Only Memory) 1011 and a RAM 1012. The ROM 1011 stores a boot program such as BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1031. The disk drive interface 1040 is connected to the disk drive 1041. For example, a removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 1041. For example, a mouse 1051 and a keyboard 1052 are connected to the serial port interface 1050. For example, a display 1061 is connected to the video adapter 1060.

  Here, as shown in FIG. 10, the hard disk drive 1031 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. Each table described in the above embodiment is stored in the hard disk drive 1031 or the memory 1010, for example.

  Further, the distributed program is stored in the hard disk drive 1031 as a program module 1093 in which a command executed by the computer 1000 is described, for example. Specifically, a program module 1093 describing each process executed by the distribution apparatus 10 described in the above embodiment is stored in the hard disk drive 1031.

  Further, data used for information processing by the distributed program is stored in the hard disk drive 1031 as the program data 1094, for example. Then, the CPU 1020 reads the program module 1093 and the program data 1094 stored in the hard disk drive 1031 to the RAM 1012 as necessary, and executes the above-described procedures.

  Note that the program module 1093 and the program data 1094 related to the distributed program are not limited to being stored in the hard disk drive 1031, but are stored in, for example, a removable storage medium and read by the CPU 1020 via the disk drive 1041 or the like. May be. Alternatively, the program module 1093 and the program data 1094 related to the distributed program are stored in another computer connected via a network such as a LAN (Local Area Network) or a WAN (Wide Area Network), and the network interface 1070 is used. It may be read by the CPU 1020.

  As mentioned above, although embodiment which applied the invention made | formed by this inventor was described, this invention is not limited with the description and drawing which make a part of indication of this invention by this embodiment. That is, other embodiments, examples, operational techniques, and the like made by those skilled in the art based on this embodiment are all included in the scope of the present invention.

DESCRIPTION OF SYMBOLS 1 Distributed database (DB) system 10 Distribution apparatus 11 Input part 12 Output part 13 Communication control part 14 Storage part 15 Control part 15a Acquisition part 15b Creation part 15c Determination part

Claims (4)

Acquire data that accompanies requests for data from each community in a database system that accommodates data belonging to the same community in the same database and distributes data belonging to different communities to multiple databases. And
A determining unit that determines a combination of communities accommodated in the same database in order of increasing communication volume between the communities among the acquired communication volumes;
A dispersion apparatus comprising:   The determining unit determines a combination of two communities as a community group in the same database in descending order of the amount of communication between the communities, and in order of increasing communication amount between the determined community groups, The distribution apparatus according to claim 1, wherein a process of determining a combination of two community groups as a community group accommodated in the same database is repeated to determine the number of community groups of the databases to be distributed. In addition, a community subgroup is created by combining a plurality of communities, and among the acquired communication amount, a creation unit that reduces variation in the entire community subgroup of the communication amount from each community subgroup,
The said determination part determines the combination of the community accommodated in the same database in order with a large communication amount between the created said community subgroups among the acquired said communication amount. Or the dispersion apparatus of 2. A distributed method executed by a distributed device,
Acquire data that accompanies requests for data from each community in a database system that accommodates data belonging to the same community in the same database and distributes data belonging to different communities to multiple databases. Process,
A determination step of determining a combination of communities accommodated in the same database in order of increasing communication volume between the communities among the acquired communication volumes;
A dispersion method characterized by comprising: