JP2007219826A - Distribution and deletion method for program, and distributed computer system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suppress a burden on a user or a program management burden on a manager when a managerial post change or the like occurs or when the user using a computer requires program distribution to the computer constituting a distributed computer system. <P>SOLUTION: A management server 101 generates agents 107, 108, sends them to the computers 109-111, and collects asset information of each the computer and attribute information of the user. An asset distribution management system produces policy information on the basis of the collected information, distributes a program required for business to each the computer inside the distributed computer system after examination/improvement by the manager, and performs deletion of the program not required for the business. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a program distribution / deletion method and a distributed computer system, and more particularly to a program distribution / deletion method and a distributed computer system for distributing / deleting a program to / from computers constituting the distributed system.

  Conventionally, in a system that distributes a program from a management server to a computer in a distributed computer system, an administrator of the management server determines which program needs to be distributed to which computer. It was necessary to perform distribution operation every time. When a large number of computers exist in the distributed computer system, performing the above-described operation for each computer is a very heavy burden on the administrator. In addition, there is a high possibility that a program is distributed to a wrong computer due to an erroneous operation by an administrator.

As a conventional technique that can solve the above-described problems, for example, a technique described in Patent Document 1 is known. This conventional technique is used when a program is distributed by a distribution information selection device. By predefining the correspondence between the transmission attributes to be transmitted and the program, the administrator can distribute the program to computers in the distributed computer system simply by specifying the program at the time of program distribution. As a result, the burden on the administrator can be reduced.
JP 2001-75811 A

  However, the above-described prior art requires the administrator to determine whether the program is necessary for business with respect to all attribute information or a program that is not necessary, and set transmission attributes for all programs. There is a problem that the work for this becomes a heavy burden on the administrator.

  The object of the present invention is to solve the problems of the prior art described above, and in order for an administrator to suppress or delete a program that is not necessary for business, it examines the target program and sets the necessity and necessity of the program one by one. An object of the present invention is to provide a program distribution / deletion method and a distributed computer system that can reduce the burden on the administrator by eliminating the necessity.

  According to the present invention, the object is to provide a program distribution and deletion method for each of the plurality of computers in a distributed computer system configured by connecting a plurality of computers and a management server. Having asset information, the management server has an attribute information table, a policy information table, a department cluster table, a position cluster table, a work form cluster table, and a distribution program, and the management server If there is a program distribution request from one, or if there is a difference in the attribute information of the user who uses the computer due to reasons such as a change in office, the user who requested the program distribution by searching the attribute information table Attribute information is extracted, and the policy information table and the department cluster table are If the policy information of the target program for the department to which the user belongs is not set in these tables, refer to the department cluster table, the position cluster table, and the work form cluster table. Select a common job system, position, and employment form from the table, generate a policy template corresponding to these, and each of the plurality of computers based on a policy that has been approved and modified by the administrator of the contents of the policy template. This is achieved by distributing and deleting programs.

  According to the present invention, it is possible to eliminate the need for an administrator to examine a target program in order to suppress or delete a program unnecessary for business, and to set the necessity and necessity of the program one by one. Can be reduced.

  Embodiments of a program distribution / deletion method and distributed computer system according to the present invention will be described below in detail with reference to the drawings.

  The present invention presupposes an environment in which a program can be installed only on each computer in the distributed computer system by distributing the program using the asset distribution management system. Therefore, in the environment of the distributed computer system using the asset distribution management system using the present invention, the program cannot be freely installed on the computer used by the user as the client.

The present invention
・ When a user requests distribution of a program from a computer,
・ If a program required for work is not installed on the computer,
・ If a computer in the distributed computer system fails and you need to reinstall all the programs you used,
・ If the computers in the distributed computer system need to be replaced due to aging, and all the programs used up to that point need to be reinstalled,
This is effective when a user changes a job system due to a department transfer or promotion, and the necessary program is distributed to a computer used by the user or an unnecessary program is deleted.

  FIG. 1 is a block diagram showing a configuration of a distributed computer system according to an embodiment of the present invention. In FIG. 1, 100 is a database, 101 is a management server, 102 is a policy determination unit, 103 is a clustering unit, 104 is a package generation unit, 105 is an agent generation unit, 106 is an agent state information management unit, 107 is agent A, 108 Is an agent B, 109 to 111 are computers, 112 is a network, 113 is an attribute information table, 114 is a policy information table, 115 is a department cluster table, 116 is a position cluster table, 117 is a work type cluster table, and 118 is agent status information. Table 119 is a destination information table, 120 is an asset information table, 121 is policy template information, and 122 is a distribution program.

  The distributed computer system according to the embodiment of the present invention shown in FIG. 1 is configured by connecting an asset distribution system including a database 100 and a management server 101 and a plurality of computers 109 to 111 via a network 112. . Although not shown, the management server 101 and the plurality of computers 109 to 111 are configured to include a storage device such as a CPU, a memory, and an HDD, an input device such as a keyboard and a mouse, and a display device. This is realized by the CPU executing the program loaded in the program.

  The database 100 includes an attribute information table 113 (details are shown in FIG. 19), a policy information table 114 (details are shown in FIG. 20), a department cluster table 115 (details are shown in FIG. 21), and a position cluster table 116. (Details are shown in FIG. 22), working mode cluster table 117 (details are shown in FIG. 23), agent status information table 118 (details are shown in FIG. 24), and destination information table 119 (details are shown in FIG. 25). ), Asset information table 120 (details are shown in FIG. 26), policy template information 121 (details are shown in FIGS. 27 and 28), and a program distributed to each computer in the distributed computer system 122 are stored and managed.

  The management server 101 includes a policy determination unit 102, a clustering unit 103, a package generation unit 104, an agent generation unit 105, an agent state information management unit 106, an agent A 107, and an agent B 108. .

  In the above description, the policy determination unit 102 is a distributed computer system when a user as a client makes a program distribution request using one of the computers 109 to 101, or when the program 122 required for business is not installed. When it is necessary to reinstall all the programs 122 used up to now when a failure occurs in the internal computer, the computer in the distributed computer system has been used up to now when it becomes obsolete and needs to be replaced. If it is necessary to reinstall all programs 122, or if there is a distribution of necessary programs 122 or deletion of unnecessary programs when a user moves or promotes a department, the policy information is restored. It is determined whether or not the program 122 can be distributed.

  The clustering unit 103 refers to the policy information set by the administrator, and creates a new cluster group if there is nothing corresponding to the cluster. Policy template information 121 is created by searching for a related department such as a department whose design is a design department such as a DB department or a system design department, and performing clustering.

  The package generation unit 104 updates the policy information of the package to the asset information table 120, and creates one or more programs 122 as a package. A package is a group of data that you want to distribute and programs that you want to install, like a parcel. After creating the package, the package generation unit 104 calls the agent generation unit 105 to create the agent B 108 and distribute the package.

  The agent generation unit 105 creates an agent A 107 and an agent B 108. The agent generation unit 105 sends the agent A 107 to all the computers in the distributed computer system to acquire asset information and client user attribute information. Further, the agent generation unit 105 distributes the program 122 by sending the agent B 108 when there is a program distribution request from the user and the policy is approved.

  When the management server 101 receives status information notifications from the agent A 107 and the agent B 108, the agent status information management unit 106 updates the status information in the agent status information table 118 (details are shown in FIG. 23).

  The agent A 107 is a program that acquires the asset information of the computer under the distributed computer system environment or the attribute information of the user as the client.

  Agent B108 moves to a computer in a distributed computer system environment and distributes the requested program 122 when a user requests distribution of the program via the computer, or when the user's own department is changed or his / her position is changed. It is a program that deletes unnecessary programs according to policy information.

  The computers 109 to 111 are computers that are used by users who are clients. In the embodiment of the present invention described below, each computer and user are linked one-to-one, and the user moves a work department. In such a case, it is assumed that the user moves with the computer used by the user. In the illustrated example, it is assumed that the computers 109 to 111 are computers that the agent A 107 has moved to acquire the asset information of the computer and the attribute information of the user. Further, it is assumed that the computer 110 is a computer that receives a program distribution request from a user as a client, and that the agent B 108 is moving to distribute the requested program 122.

  FIG. 2 is a block diagram showing the configuration of the agent A 107. The agent A 107 includes an asset information collection unit 200, an asset information registration unit 201, an attribute information collection unit 202, and an attribute information registration unit 203. The asset information collection unit 200 of the agent A 107 collects the asset information of the computer under the distributed computer system environment and provides the collected asset information to the asset information registration unit 201. Details of the asset information to be collected will be described later with reference to FIGS. The asset information registration unit 201 of the agent A 107 collects the asset information of computers under the distributed computer system environment, and registers the asset information collected in the database 100 after the agent A 107 returns to the management server 101. Further, the agent A 107 collects attribute information from a user who uses the computer under the distributed computer system environment and provides it to the attribute information registration unit 203. Details of the attribute information to be collected will be described later with reference to FIGS. The attribute information registration unit 203 of the agent A 107 collects attribute information from users of computers under the distributed computer system environment, and registers the collected attribute information in the database 100 after the agent A 107 returns to the management server 101.

  FIG. 3 is a block diagram showing the configuration of the agent B108. The agent B 108 includes a destination information management unit 300, a package management unit 301, a program distribution / deletion execution unit 302, and a task completion notification unit 303. The destination information management unit 300 of the agent B 108 manages a destination information table 119 (details are shown in FIG. 25) of the computer that the agent B 108 goes to distribute or delete the program 122. The agent B 108 refers to the destination information table 119, moves to the target computer, distributes the program 122 necessary for the business, and deletes the program unnecessary for the business. The package management unit 301 of the agent B 108 is a place for storing several programs 122 (packages) requested by the user via the computer. The program distribution / deletion execution unit 302 of the agent B 108 moves when the agent B 108 moves to a computer that has requested the program distribution or a computer that has changed the program 122 that can be used by the client user due to a change in office. It is an execution unit that performs the distribution process of the program 122 necessary for the business or the deletion process of the program unnecessary for the business. The task completion notifying unit 303 of the agent B 108 executes a function of notifying the management server 101 when the agent B 108 completes the distribution processing of the program 122 necessary for the business or the deletion processing of the program unnecessary for the business. .

  FIG. 4 shows the overall processing operation of the embodiment of the present invention until the distribution of the program 122 required for the business based on the policy information or the deletion of the program unnecessary for the business is completed for the computers in the distributed computer system. This will be described next.

(1) First, the policy information of the program 122 that the administrator can use for each department / position / work type is set and registered in the asset management system (step 400).

(2) After the policy information is set, the management server 101 generates an agent A 107 (original agent) that acquires asset information or user attribute information in all computers under the distributed computer system environment and a clone agent of the agent A 107 To do. Details of this processing will be described later with reference to FIG. 15 (step 401).

(3) Next, the management server 101 transmits the clone agent of the agent A 107 generated in the process of step 401 to each computer (here, the computer 109 and the computer 110) in the distributed computer system environment. Details of this processing will be described later with reference to FIG. 16 (steps 402 and 403).

(4) The clone agent of the agent A 107 moved to the computer 109 and the computer 110 acquires the asset information and user attribute information of the computer 109 and the computer 110. Details of this processing will be described later with reference to FIG. 16 (steps 404 and 406).

(5) In addition, the clone agent of the agent A 107 moved to the computer 109 and the computer 110 transmits the acquired asset information and user attribute information to the original agent in the management server 101 by agent communication. Details of this processing will be described later with reference to FIG. 16 (steps 405 and 408).

(6) Next, the original agent of the agent A 107 of the management server 101 stores the asset information and user attribute information of each computer in the distributed computer system environment transmitted in the processes of steps 405 and 408 in the database 100. Then, the management server 101 is notified of the processing result. Details of this processing will be described later with reference to FIGS. 16 and 17 (steps 407 and 409).

  The management server 101 issues a request to move the agent A 107 to the agent generation unit 105 at regular intervals, and executes the above-described processing for acquiring asset information and user attribute information from a computer under the distributed computer system environment. To do. Details of this processing will be described later with reference to FIGS.

(7) Next, the management server 101 extracts the asset information and user attribute information of each computer under the distributed computer system environment from the database 100 and generates policy template information 121. After the policy template information 121 is generated, a policy confirmation request is issued to the administrator. The administrator approves if there is no need for correction, changes it if necessary, and registers the policy template information 121 as policy information. Details of this processing will be described later with reference to FIGS. 5 to 9 and FIGS. 11 to 14 (step 410).

(8) Next, it is assumed that there is a program distribution request from the computer 110 or a change notification of attribute information of a user as a client (step 411).

(9) Upon receiving the request or notification in step 411, the management server 101 generates a package for distributing the requested program 122. Details of this processing will be described later with reference to FIG. 10 (step 412).

(10) Next, the management server 101 generates an agent B108 (step 413), issues a movement request to the generated agent B108, and moves the agent B108 to the computer 110. Details of this processing will be described later with reference to FIGS. 10, 15, and 18 (steps 413 to 415).

(11) The agent B 108 distributes a package necessary for business or deletes a program unnecessary for business in the migration destination computer 110. Details of this processing will be described later with reference to FIG. 18 (step 416).

(12) Next, the agent B 108 notifies the management server 101 of the processing result, notifies the management server 101 of the processing result, and then disappears. Details of this processing will be described later with reference to FIGS. 17 and 18 (steps 417 and 418).

  The series of processing from step 400 to step 418 described above is repeated unless there is a forced termination by the administrator or an abnormal termination of the system.

  FIG. 5 is a flowchart for explaining the processing operation in the policy determination unit 102, which will be described next.

(1) First, the administrator sets policy information. After that, through the process 0 indicating the loop point returned from the process flow described later, the agent generation unit 105 is called to generate the agent A107. After the generation of the agent A 107, the agent A 107 is moved to all computers in the distributed computer system, and the asset information in each computer and the attribute information of the user as a client are collected (steps 500 to 502).

(2) Next, the attribute information table 113 (FIG. 19) stored in the database 100 is referred to. The attribute information table 113 stores and manages user attribute information collected by the agent A 107 in all computers in the distributed computer system (step 503).

(3) Next, it is checked whether there is a difference from the previously checked information. If there is a change in the attribute information, the process proceeds to a difference information extraction process (process 1) described later with reference to FIG. 6 (step 504). 505).

(4) If there is no change in the attribute information in the check in step 504, it is checked whether or not a program distribution request is received from a user as a client. If there is no program distribution request, the processing from step 503 is performed. If there is a program distribution request, the process proceeds to attribute information extraction processing (processing 2) (steps 506 and 507).

  FIG. 6 is a flowchart for explaining the operation of the difference information extraction process (Process 1) at Step 505 in FIG. 5, which will be described next.

(1) First, the difference between the attribute information of the client user collected by the agent A 107 last time and the attribute information of the client user collected by the agent A 107 this time is extracted (step 600).

(2) Next, a program distribution permission determination process (process 3) described later is executed, and if there is a difference in the attribute information of the user due to reasons such as a change in office, etc., a program unnecessary for business using the agent B108 Is deleted, and it is determined whether or not the deletion processing of the program unnecessary for the business is successful (steps 601 and 602).

(3) If it is determined in step 602 that the program unnecessary for the business has not been successfully deleted, the process returns to the process from step 602, and the deletion process is continued to delete the program unnecessary for the business. If successful, the asset information in the database 100 is updated (step 603).

(4) After updating the asset information in the database 100, the administrator is notified that the update of the asset information has been completed, and then the process proceeds to process 0 (steps 604 and 605).

  FIG. 7 is a flowchart for explaining the operation of the attribute information extraction process (process 2) at step 507 in FIG. 5. Next, this will be described.

(1) When there is a program distribution request from a user as a client or when there is a difference in attribute information due to reasons such as changing the office system, the attribute information table 114 in the database 100 is searched (step 700).

(2) Next, the attribute information of the user who requested the program distribution is extracted, and then the process proceeds to the program distribution permission determination process (process 3) in step 601 in FIG. 6 (steps 701 and 702).

  FIG. 8 is a flowchart for explaining the operation of the program distribution permission determination processing (processing 3) in steps 601 and 702 in FIGS. 6 and 7, which will be described next.

(1) First, referring to the policy information table 114 (described later with reference to FIG. 20) and the department cluster table 115 (described later with reference to FIG. 21) in the database 100, the target for the department to which the client user belongs is stored in these tables. It is checked whether or not the policy information of the program 122 to be set is set (steps 800 and 801).

(2) If the policy of the program is not set in the check in step 801, the process proceeds to a clustering process (process 4) described later. If the policy information is set, the program 122 is instructed to the computer. It is checked whether or not distribution is permitted (steps 805 and 802).

(3) If distribution of the program 122 is permitted in the check in step 802, the process proceeds to package generation processing (Process 5) described later. If distribution of the program 122 is not permitted, the user who uses the computer In response to this, a rejection notification that distribution of the program 122 is not permitted is issued (steps 804 and 803).

  FIG. 9 is a flowchart for explaining the operation of the clustering process (process 4) in step 805 in FIG. 8, and this will be described next.

(1) Since the policy of the target program 122 is not set for the department to which the client user belongs in the check in step 801 in FIG. 8, first, the department cluster table 115 is referred to, It is checked whether or not there is information (design department, sales department, etc.), and if there is common job type information, the process proceeds to a department cluster selection process (process 6) described later (steps 900 and 901).

(2) If there is no common job type information in the check in step 900, a new department cluster is created to check whether there is common job system information. If there is common job system information, it will be described later. The process shifts to a position cluster selection process (process 7) (steps 902 to 904).

(3) If there is no common job system information in the check in step 903, a new job system cluster is created, and then it is checked whether there is common job type information (steps 905 and 906).

(4) If there is common work form information in the check in step 906, the process shifts to a work form cluster selection process (Process 8) described later. If there is no common work form information, a new work form cluster is created. Is created, and the process proceeds to step 6 in step 901 (steps 907 and 908).

  FIG. 10 is a flowchart for explaining the operation of the package generation process (process 5) at step 804 in FIG. 8, which will be described next.

(1) In the check at step 802 in FIG. 8, the policy of the target program 122 is set for the department to which the user as the client belongs, and the distribution of the program 122 is permitted to the computer. First, the package generation unit 104 is called (step 1000).

(2) The program 122 to be registered in the package is selected, and then a department cluster, a position cluster, and a work form cluster that can acquire the package are set (steps 1101 and 1002).

(3) Next, after registering package information in the policy information table 114 in the database 100 and creating a package, the process proceeds to an agent generation process (process 10) described later with reference to FIG. (Steps 1003 and 1005).

(4) After generating the agent B108 by the process 10, the agent B108 is moved to the computer that requested the program distribution or the computer that needs to be deleted. After agent B 108 moves to the computer that requested the program distribution, it notifies whether or not the distribution of the package or the deletion of the program unnecessary for the business was successful. If the check is unsuccessful, the process returns to the process from the process 10 in the step 1005 to repeat the process (step 1006).

(5) If the process of distributing the package or deleting the unnecessary program is successful in the check of step 1006, the process is notified to the administrator and the process proceeds to process 0 (steps 1007 and 1008).

  FIG. 11 is a flowchart for explaining the operation of the department cluster selection process (process 6) at step 901 in FIG. 9. Next, this will be described.

(1) The policy of the target program 122 is not set for the department to which the user who is the client belongs, and there is common job type information (design department, sales department, etc.) in the department cluster table 115. Then, it is checked whether or not there is a plurality of department information in the department cluster table 115 (details are shown in FIG. 21) (step 1100).

(2) If there is a plurality of department information in the check in step 1100, the department cluster with the largest number of components is selected. If there are department clusters with the same number of components, the department cluster with the smallest registration number is selected. Is selected (step 1101).

(3) After the process of step 1101 or when there is no plurality of department information in the check of step 1100, the process proceeds to a policy information creation process (process 9) described later (step 1102).

  FIG. 12 is a flowchart for explaining the operation of the position cluster selection process (process 7) at step 904 in FIG. 9, which will be described next.

(1) Since the policy of the target program 122 is not set for the department to which the client user belongs in the check in step 903 in FIG. 9, and there is common job system information in the department cluster table 115, It is checked whether or not there are a plurality of position information (step 1200).

(2) If there are a plurality of position information in the check in step 1200, the position cluster with the largest number of constituents is selected, and if there are position clusters with the same number of constituents, the position cluster with the smallest registration number is selected. Is selected (step 1201).

(3) After the process of step 1201 or when there is no plural position information in the check of step 1200, the process proceeds to a policy information creation process (process 9) described later (step 1202).

  FIG. 13 is a flowchart for explaining the operation of the work form cluster selection process (process 8) at step 907 in FIG. 9. Next, this will be described.

(1) In the check at step 906 in FIG. 9, the policy of the target program 122 is not set for the department to which the user as the client belongs, and the department cluster table 115 has a common work form. Thus, it is checked whether or not there is a plurality of employment form information (step 1300).

(2) If there are a plurality of employment type information in the check in step 1300, the employment type cluster having the largest number of components is selected, and if there are employment type clusters having the same number of components, the highest registration number A small work form cluster is selected (step 1301).

(3) After the processing in step 1301 or when there is no plurality of work form information in the check in step 1300, the process proceeds to policy information creation processing (processing 9) (step 1302).

  FIG. 14 is a flowchart for explaining the operation of the policy information creation processing (processing 9) in steps 1102, 1202, and 1302 in FIGS. 11 to 13. This will be described next.

(1) When processing is started, first, policy template information 121 is created based on new policy information that combines the selected department cluster, occupation cluster, and employment form cluster (step 1400).

(2) After the policy template information 121 is created in the process of step 1400, this is notified to the administrator, and the content of the policy template information 121 is confirmed (step 1401).

(3) It is checked whether or not the policy information is approved by the administrator. If the policy is not approved by the administrator, the administrator himself corrects the contents of the policy template information 121 (steps 1402 and 1405).

(4) If the administrator approves the check in step 1402, the policy template information generated in the process of step 1400 is corrected. If the administrator corrects the policy template in the process of step 1405, Information is updated in the policy information table 114 of the database 100 (details will be described with reference to FIG. 20), and the program distribution permission determination process (process 3) described above is performed (steps 1403 and 1404).

  FIG. 15 is a flowchart for explaining the operation of the agent generation process (process 10) in step 1005 in FIG. 10, and this will be described next.

(1) When the processing is started, first, the agent generation unit 105 is called to check whether or not a request for creating the agent A 107 is received from the package generation unit 104 (steps 1500 and 1501).

(2) If a request to create agent A107 is received in the check in step 1501, agent A107 is generated, and asset information table 120 (details will be described with reference to FIG. 26) is mounted on the generated agent A107. And a destination information table 119 (details will be described with reference to FIG. 25) (steps 1502 and 1503).

(3) Further, the attribute information table 113 (details will be described with reference to FIG. 19) is mounted on the generated agent A 107 to create a clone agent of the agent A 107 (steps 1504 and 1505).

(4) Thereafter, a move request is issued to the clone agent of the agent A 107, and after the clone agent of the agent A 107 moves, the process proceeds to an agent state information management process (process 11) described later with reference to FIG. It is notified that it is moving (steps 1506 and 1507).

(5) If the agent A107 creation request has not been received from the package generation unit 104 in the check in step 1501, it is checked whether or not the agent B108 creation request has been received, and the agent B108 creation request is received. If not, the process returns to step 1501 and continues (step 1508).

(6) If the agent B108 creation request has been received in the check in step 1508, the agent B108 is generated, the already generated package is mounted on the generated agent B108, and the asset information table 120 (details are shown in FIG. 26 (steps 1509 and 1510).

(7) Further, the destination information table 119 (details will be described with reference to FIG. 25) is mounted on the generated agent B 108, and a movement request is issued to the agent B 108 (steps 1511 and 1512).

(8) After the agent B 108 moves, the process proceeds to an agent state information management process (process 11) described later with reference to FIG. 17, and notifies that the agent B 108 is moving (step 1513).

  FIG. 16 is a flowchart for explaining the processing operation of the agent A 107. Next, this will be explained.

(1) After being generated, the agent A 107 checks whether or not a movement request has been received from the agent generation unit 105, and if not received, waits until it is received (step 1600).

(2) When a move request is received from the agent generation unit 105 in the check in step 1600, the destination information table 119 (details are shown in FIG. 25) is referred to, and the clone agent of the agent A 107 created in the request destination computer is selected. Move (step 1601).

(3) The clone agent resides in the destination computer after arrival and collects the asset information of the computer and the attribute information of the client user. At this time, an attribute information setting notification is sent to the user as a client to set the attribute information (step 1602).

(4) The clone agent of the agent A 107 collects the asset information of the computer and the attribute information of the client user, and then transmits the information collected by agent communication to the original agent in the management server 101 (step 1603). .

(5) After the original agent receives the information from the clone agent, the asset information table 120 (FIG. 26) and the attribute information table in which the collected asset information of the computer and the attribute information of the client user are stored in the database 100. 113 (FIG. 19) and the table is updated (step 1604).

(6) Next, the process proceeds to the agent state information management process (process 11) to notify the completion of the process in the agent A 107, and then returns to the process from step 1602 at regular intervals, and the process described above Is repeated (step 1605).

  FIG. 17 is a flowchart for explaining the operation of the agent state information management process (Process 11) in Steps 1507 and 1513 in FIG. 15 and Step 1605 in FIG. 16, which will be described next.

(1) When processing is started, the agent status information management unit 106 is first called to check whether status information notification is received from the agent A 107 or the agent B 108 (steps 1700 and 1701).

(2) When the status information notification is received in the check in step 1701, the agent status information table 118 (details will be described with reference to FIG. 24) of the database 100 is updated with the received information (step 1702).

(3) If the status information notification from agent A 107 or agent B 108 does not come within a certain time in the check in step 1701, it is considered that agent A 107 or agent B 108 has disappeared for some reason, and a clone agent is created. Then, the created clone agent is moved to the computer with the disappeared agent (steps 1703 and 1704).

  FIG. 18 is a flowchart for explaining the processing operation of the agent B 108. Next, this will be explained.

(1) After being generated, the agent B 108 checks whether or not a movement request has been received from the agent generation unit 105. If not received, the agent B 108 waits until it is received (step 1800).

(2) When the movement request is received from the agent generation unit 105 in the check in step 1800, the agent B 108 refers to the destination information table 119 (details are shown in FIG. 25) and moves to the requested computer (step 1801).

(3) The agent B 108 distributes the package in the destination computer, and after distributing the package, refers to the asset information table 120 (details are shown in FIG. 26) and confirms whether there is a program to be deleted. (Steps 1802 and 1803).

(4) If it is determined in step 1803 that there is no program to be deleted, the process proceeds to the already described agent state information management process (process 11), and the agent B 108 notifies the process state (step 1804).

(5) Next, it is checked whether or not the state of the agent is processing complete. If the processing is complete, the agent B 108 disappears (steps 1805 and 1806).

(6) If it is determined in step 1805 that the process has not been completed, it is determined that the agent has disappeared for some reason. (The agent that has already disappeared because the clone agent created in process 11 has performed package distribution) Returning to the processing from step 1802, the package distribution is continued.

(7) After the package is distributed in the check in step 1803, the asset information table 120 (details are shown in FIG. 26) is referred to, and if there is a program that is unnecessary for business (to be deleted), the program is deleted. It is checked whether or not the process of deleting the program unnecessary for the business is successful (steps 1807 and 1808).

(8) If the check in step 1808 succeeds in deleting the program unnecessary for the job, the process proceeds to the agent status information management process (step 11) already described in step 1804 to delete the program unnecessary for the job. If the process has failed, the process returns to step 1807 to continue the program deletion process.

  FIG. 19 is a diagram illustrating a configuration example of the attribute information table 113. The attribute information table 113 manages user attribute information as a client using each computer in the distributed computer system environment. The attribute information managed here is name information, department information, job system information, and employment form information. When the management server 101 distributes a program 122 necessary for business or deletes a program unnecessary for business. To be referenced.

  FIG. 20 is a diagram illustrating a configuration example of the policy information table 114. The policy information table 114 is permitted to install a package name, a program name (a program 122 that the administrator determines to be necessary for business with respect to each computer in the distributed computer system and distributed), and a program. Manages the target policy information (combination of department cluster, position cluster, and work type cluster).

  The asset distribution management system refers to the information in the policy information table 114 and the information in the attribute information table 113, and if all the conditions of the department information, the organization information, and the employment form information in the attribute information match the policy information, the program Distribute or delete As used herein, a cluster refers to a collection of data obtained by grouping together data defined as being similar by an administrator into one group and dividing one large set of data into a small set of data.

  An example of the policy information table 114 before the clustering process of the asset distribution management system is shown in FIG. 20A. As a first example in this example, the package 1 in which the program A, the program B, and the program C are packaged. The policy information that can be installed by the quality assurance section (A10), dispatched or normal (C3), and in charge (B1) is set. As a second example, the package 2 in which the program D and the program E are packaged is installed by the DB section (A2), dispatch or normal (C3), and the chief or section manager (B13). Policy information that can be set. FIG. 20B shows an example of the policy information table 114 after the clustering process of the asset distribution management system is performed. As a third example in this example, the program D and the program E are packaged. The package 2 has been changed to policy information that can be installed by the DB department or the system design department (A11), dispatched or normal (C3), and the chief or section manager (B13). Show.

  FIG. 21 is a diagram showing a configuration example of the department cluster table 115. The department cluster table 115 manages the department cluster name composed of a single department or a plurality of departments, the target department signature, the department type, and the number of data constituting the department cluster (number of components).

  FIG. 22 is a diagram showing a configuration example of the position cluster table 116. The position cluster table 116 manages a position cluster name composed of a single or a plurality of positions, a target (position) name, and the number of data (number of components) constituting the position cluster.

  FIG. 23 is a diagram showing a configuration example of the work form cluster table 117. The employment form cluster table 117 manages an employment form cluster name composed of one or a plurality of employment forms, a target (working form) name, and the number of data (configuration number) constituting the employment form cluster.

  FIG. 24 is a diagram showing a configuration example of the agent state information table 118. The agent status information table 118 includes names of agents that collect computer asset information and client user attribute information, names of agents that distribute programs 122 required for business operations or delete programs that are unnecessary for business operations, and agent statuses. And manage.

  FIG. 25 is a diagram illustrating a configuration example of the destination information table 119. The destination information table 119 manages the agent name, the computer name to which the agent moves, and its IP address.

  FIG. 26 is a diagram illustrating a configuration example of the asset information table 120. The asset information table 120 manages computer names, program names, user names, and deletion targets of programs unnecessary for business as asset information of computers in a distributed environment.

  FIG. 27 is a diagram showing the relationship between job types, department clusters, and programs before clustering processing when creating a policy template, and FIG. 28 is a diagram showing the relationship between job types, department clusters, and programs after clustering processing when creating a policy template. Next, with reference to FIGS. 27 and 28, an example of a change in the relationship among job types, department clusters, and programs before and after the clustering process will be described.

  Now, assume that agent A 107 collects asset information from computers in the distributed computer system, and the attribute information, which is one of the asset information, includes information such as the system design department and the supervisor. Since nothing is set in the policy of the system design section, the arrow connecting the system design section and the program is not shown in FIG. Since the system design unit is the design department, the clustering unit 103 searches the department cluster table 113 of the database for a design department that has already been registered. In the example shown in FIG. 27, policy information for the DB unit and the quality assurance unit has already been set, and an arrow to the program is shown.

  As shown by the connected line, the policy information of the DB section includes a section manager and a chief manager as the target job system, and dispatch and normal are targeted as work modes. The DB unit is a target program that can be used by the program D and the program E indicated by the arrows from the DB unit. The policy information of the Quality Assurance Department is in charge of the target job system as shown by the connected line, and the employment type is targeted for dispatch and normal. The target programs that can be used by the quality assurance unit are the target programs that can be used by the programs A, B, and C indicated by arrows from the quality assurance unit.

  At this point, the policy information of the system design unit is not set, but since it is a design unit, the clustering unit 103 infers that the system design unit belongs to the same design cluster as the DB unit. In the policy information of the DB section, since the section manager and the person in charge of the manager are permitted to use the program C and the program D, the clustering section 103 displays the section manager position of the system design section as shown by a dotted line in FIG. Policy information that can be used by the person in charge of the person in charge of the person in charge and the person in charge of the supervisor, and as shown by the dotted arrows in FIG. 28, the program D and the program E are set as target programs that can be used.

  In the above-described example, policy information that can distribute a program is set. However, policy information that cannot distribute a program 122 can also be set. In addition, the proposed policy information includes department information, position information, and employment type. In addition to this, address information can also be cited as an example of a judgment criterion.

  FIG. 29 is a diagram for explaining a processing example in the embodiment of the present invention when a user attribute is changed.

  In FIG. 29, it is assumed that Mr. A has been transferred from the system design department to the information system department. In this case, the asset distribution management system detects that the department to which Mr. A belongs has been changed based on the asset information of the computer collected by the agent. Then, the asset distribution management system refers to the policy information table and searches for the corresponding policy information. After searching the corresponding policy, an installable program and differential program information to be deleted are extracted, and a necessary program distribution process and an unnecessary program deletion process are performed.

  In addition, it is assumed that the computer used by Mr. B has failed and cannot be restored, and the OS has been replaced. In this case, the asset distribution management system detects that the software required for business is not contained in Mr. B's computer based on the asset information of the computer collected by the agent, refers to the policy information table, and then applies the corresponding policy. Search for information. Then, the asset distribution management system performs distribution processing of installable programs after searching for the corresponding policy information.

  In addition, it is assumed that Mr. C has become the chief manager of the job system information due to the promotion. In this case, the asset distribution management system refers to the policy information table and searches for the corresponding policy information. Then, the asset distribution management system extracts the installable program and the difference program information to be deleted after searching the corresponding policy, and performs the distribution process of the necessary program and the deletion process of the unnecessary program.

  In the above description, the attribute change may be registered by each user in the computer used by the user, or may be automatically registered from a system that manages personnel and the like (not shown).

  Each process in the above-described embodiment of the present invention is configured by a program, and can be executed by a CPU provided in the management server, and these programs are stored in a recording medium such as an FD, a CD-ROM, or a DVD. It can also be provided as digital information via a network.

  According to the above-described embodiment of the present invention, the asset distribution management system can increase the accuracy of inference in the system as the administrator's judgment information is accumulated in the database, thereby reducing the burden on the administrator. it can. In addition to reducing the burden on the administrator, it is possible to speed up and increase the efficiency of work for client users.

  As described above, the asset distribution management system used in the embodiment of the present invention collects asset information of a computer or user attribute information in a distributed computer system using an agent, and infers based on the attribute information. Based on the policy information, the program is distributed or deleted. As a result, the administrator only needs to correct the policy information only when there is an error in the policy information inferred by the asset distribution system, and investigates whether or not the program is a program necessary for business. It is possible to reduce the burden on the administrator who distributes or deletes programs to the client computers in the distributed computer system.

  In the embodiment of the present invention, the administrator defines policy information of a program necessary for business or a program unnecessary for business for attribute information that should be set in advance (for example, on the setting screen). In this case, it is only necessary to set the policy information by defining the program information to be distributed by combining the cluster information such as the department cluster, the position cluster, and the employment form cluster. After that, the asset distribution management system associates the attribute information not yet set by the administrator based on the asset information of the computer and the attribute information of the client user, and generates policy template information of the installable program. . Administrators only need to modify the generated policy template information. As judgment information related to administrator policy information accumulates in the database, the inference system in the system increases, reducing the burden on the administrator. can do. Further, not only can the burden on the administrator be reduced, but also the work can be speeded up and made more efficient for the user.

  For example, if there is a computer on which a program required for business is not installed, the asset distribution management system installs the program to the computer based on the policy information even if the client user does not issue a program distribution request. Can be distributed. In addition, if a computer in a distributed computer system cannot be restored due to a failure or if it becomes necessary to replace the computer due to aging, all the programs that have been used up to that point will be deleted. Even in such a case, in the embodiment of the present invention, in the embodiment of the present invention, the asset distribution management system performs the calculation based on the policy information such as the job type, department, job title, and work format to which the client user belongs. Since the program is distributed to the user, the burden of environment construction by the user can be reduced.

It is a block diagram which shows the structure of the distributed computer system by one Embodiment of this invention. 4 is a block diagram showing a configuration of agent A. FIG. 4 is a block diagram showing a configuration of an agent B. FIG. It is a flowchart explaining the whole processing operation of embodiment of this invention until distribution or deletion of the program with respect to the computer in a distributed computer system is completed. It is a flowchart explaining the processing operation in a policy determination part. 6 is a flowchart for explaining an operation of a difference information extraction process (Process 1) in Step 505 in FIG. It is a flowchart explaining the operation | movement of the attribute information extraction process (process 2) in step 507 in FIG. 8 is a flowchart for explaining the operation of program distribution permission determination processing (processing 3) in steps 601 and 702 in FIGS. 6 and 7. It is a flowchart explaining the operation | movement of the clustering process (process 4) in step 805 in FIG. FIG. 9 is a flowchart for explaining an operation of package generation processing (processing 5) in step 804 in FIG. 8; FIG. 10 is a flowchart for explaining the operation of department cluster selection processing (processing 6) at step 901 in FIG. FIG. 10 is a flowchart for explaining an operation of a position cluster selection process (process 7) in step 904 in FIG. 9; FIG. It is a flowchart explaining the operation | movement of the employment form cluster selection process (process 8) in step 907 in FIG. 14 is a flowchart for explaining an operation of policy information creation processing (processing 9) in steps 1102, 1202, and 1302 in FIGS. It is a flowchart explaining the operation | movement of the agent production | generation process (process 10) in step 1005 in FIG. It is a flowchart explaining the processing operation of Agent A107. 17 is a flowchart for explaining the operation of agent state information management processing (processing 11) in steps 1507 and 1513 in FIG. 15 and in step 1605 in FIG. It is a flowchart explaining the processing operation of Agent B108. It is a figure which shows the structural example of an attribute information table. It is a figure which shows the structural example of a policy information table. It is a figure which shows the structural example of a department cluster table. It is a figure which shows the structural example of a position cluster table. It is a figure which shows the structural example of an employment form cluster table. It is a figure which shows the structural example of an agent state information table. It is a figure which shows the structural example of a destination information table. It is a figure which shows the structural example of an asset information table. It is a figure which shows the relationship between the job type before a clustering process at the time of policy template creation, a department cluster, and a program. It is a figure which shows the mutual relationship between the job type after a clustering process at the time of policy template preparation, a department cluster, and a program. It is a figure explaining the process example in embodiment of this invention when a user's attribute is changed.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Database 101 Management server 102 Policy determination part 103 Clustering part 104 Package generation part 105 Agent generation part 106 Agent state information management part 107 Agent A
108 Agent B
109 Computer 110 Computer 111 Computer 112 Network 113 Attribute Information Table 114 Policy Information Table 115 Department Cluster Table 116 Position Cluster Table 117 Work Cluster Table 118 Agent Status Information Table 119 Destination Information Table 120 Asset Information Table 121 Policy Template Information 122 Distribution Program

Claims (5)

In a program distribution and deletion method for each of the plurality of computers in a distributed computer system configured by connecting a plurality of computers and a management server,
Each of the plurality of computers has asset information, and the management server has an attribute information table, a policy information table, a department cluster table, a position cluster table, an employment form cluster table, and a distribution program,
The management server searches the attribute information table when there is a program distribution request from one of the computers, or when there is a difference in attribute information of a user who uses the computer due to reasons such as a change in office. Then, the attribute information of the user who requested the program distribution is extracted, the policy information table and the department cluster table are referred to, and the policy information of the target program is set for the department to which the user belongs in these tables. If not, refer to the department cluster table, position cluster table, and work form cluster table, select a common position system, position, and work form from these tables, generate a policy template corresponding to these, Policy template content based on a policy that has been approved and modified by the administrator There are, distribute programs to each of said plurality of computers, distribution of the program and performs the deletion, delete.   When selecting a job system from the department cluster table, if there are multiple common job type information in the department cluster table, select the department cluster with the largest number of constituents, and there are department clusters with the same number of constituents. 2. The program distribution / deletion method according to claim 1, wherein the department cluster having the smallest registration number is selected.   When selecting a position from the position cluster table, if there are multiple common position information in the position cluster table, the position cluster with the largest number of positions is selected, and the same number of position clusters exist. 3. The program distribution / deletion method according to claim 1 or 2, wherein the position cluster having the smallest registration number is selected.   When selecting a work form from the work form cluster table, if there are a plurality of common work form information in the work form cluster table, select the work form cluster with the largest number of constituents, and work forms with the same number of constituents. 4. The program distribution / deletion method according to claim 1, wherein when the cluster exists, an employment form cluster having the smallest registration number is selected. In a distributed computer system in which a plurality of computers and a management server are connected, and the management server distributes and deletes a program to each of the plurality of computers,
Each of the plurality of computers has asset information, and the management server has an attribute information table, a policy information table, a department cluster table, a position cluster table, an employment form cluster table, and a distribution program,
The management server searches the attribute information table when there is a program distribution request from one of the computers or when there is a difference in the attribute information of a user who uses the computer due to reasons such as a change in office. Then, the attribute information of the user who requested the program distribution is extracted, the policy information table and the department cluster table are referred to, and the policy information of the target program is set for the department to which the user belongs in these tables. If not, refer to the department cluster table, position cluster table, and work form cluster table, select a common position system, position, and work form from these tables, generate a policy template corresponding to these, Based on policies approved and modified by the administrator of policy template contents There are, distribute programs to each of said plurality of computers, distributed computer system characterized by performing the deletion.

Images