JP2017049939A - Computer system and data distribution method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically distribute data.SOLUTION: A computer system for distributing data for managing a job is configured by connecting a plurality of computers for executing jobs over a network. Each of the computers includes a processor for executing a program, and a storage device for storing the program. One of the computers collects addresses of computers which are to receive the data, in a predetermined range of the network, and distributes the data to the computers of the addresses collected.SELECTED DRAWING: Figure 6

Description

  The present invention relates to a computer system including a plurality of computers that execute jobs, and more particularly to a method for distributing data for managing jobs.

  In client / server systems, the performance of personal computers, tablet PCs, smartphones, etc. used as client machines has been improved, and there is a need to automatically execute business programs on client machines and to manage operation of client machines. It is growing.

  On the other hand, job management systems are used in order to schedule and automatically execute daily business programs. The job management system manages the business programs to be executed in units of jobs, and specifies various processing cycles such as daily or monthly calendars that set operating days or closed days, so that it can be adapted to various tasks. You can schedule jobs.

  A conventional system includes a manager machine and an agent machine, and an administrator operates the manager machine to instruct the agent machine on the job execution order and execution timing. Further, the agent machines can be linked by sequentially issuing job execution instructions from the manager machine to each agent machine.

  As background arts in this technical field, there are JP-A-2008-90485 (Patent Document 1) and JP-A-2008-252297 (Patent Document 2).

  In Patent Document 1, a job registration receiving unit receives job registration information including a job definition, which is information necessary for job execution, and a batch command executed according to the job in an information system having a plurality of servers. The job management unit receives and holds the approved job registration information from the job registration receiving unit. The distribution unit distributes the batch command in the job registration information to the servers in the information system. The job definition holding unit holds a job definition in the job registration information. The matching unit executes matching between the job registration information held in the job management unit, the information related to the batch command distributed to the server, and the job definition held in the job definition holding unit, and the batch command and job definition Describes a job management apparatus that determines whether or not the two match.

  Further, in Patent Document 2, the network configuration information collection processing unit transmits a Ping command to each device of the network to be managed, and transmits a Ping command in which the transmission source IP address is rewritten, and the response By performing the above, communication is generated between the network devices, and the communication history information of each network device is collected. Thus, a network management system is described in which bidirectional communication is generated between network devices by communication from a single point, and configuration information including a connection configuration between network devices is collected without depending on randomness.

JP 2008-90485 A JP 2008-252297 A

  In the above-described conventional job management system, the manager machine manages all agent machines and issues job execution instructions. Further, the administrator needs to individually set the operation of each agent machine, and the management cost increases as the number of machines increases. Due to such machine performance and cost constraints, it is difficult to operate a job management system including a large number of machines exceeding several thousand to several tens of thousands.

  Therefore, an autonomous distributed job management system that realizes a single large system by automatically executing jobs on a single machine without using a manager machine and linking the machines to each other is desired. ing. In order to realize an autonomous distributed job management system, it is necessary to distribute a file in which setting information such as a job and a schedule is defined (hereinafter, a definition file) to each machine.

  However, the administrator manages the machine to which the definition file is distributed in advance, and manually copies the definition file to each machine, or registers the distribution machine in the file transfer software and distributes the definition file. However, with this method, it is difficult to distribute definition files to a large number of machines exceeding thousands to tens of thousands. In addition, there is a problem that it is difficult to cope with a newly increased number of machines and a change of machines due to a failure. In addition to the work of creating a job definition, the administrator needs to perform the work of creating a distribution definition separately, which increases the amount of work.

  An object of the present invention is to automatically identify a machine on which a job management program is installed and automatically distribute a file even when an administrator cannot manage or register a distribution destination machine of a definition file.

  A typical example of the invention disclosed in the present application is as follows. That is, a computer system that distributes data for managing a job, wherein the computer system is configured by connecting a plurality of computers that execute jobs via a network, and each of the plurality of computers is A processor that executes the program and a storage device that stores the program, and one of the plurality of computers collects addresses of computers to which the data is distributed within a predetermined range of the network. The data is distributed to the computers having the collected addresses.

  According to the exemplary embodiment of the present invention, it is possible to dynamically cope with changes in the system configuration such as addition / deletion of machines. Problems, configurations, and effects other than those described above will become apparent from the description of the following embodiments.

It is a block diagram of the job management system of 1st Example. It is a block diagram of the autonomous distributed type job management program of 1st Example. It is a block diagram which shows the physical structure of the distribution destination machine of 1st Example. It is a figure explaining preparation of the destination list of the 1st example. It is a flowchart of the process which produces the destination list | wrist of 1st Example. It is a figure explaining distribution of the definition file of 1st Example. It is a flowchart of the process which distributes a definition file to the destination produced | generated from the dependency of the machine acquired from the job definition of 2nd Example. It is a flowchart of the process which finds the machine which has the job management program of 2nd Example.

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

<Example 1>
FIG. 1 is a configuration diagram of a job management system according to the first embodiment of this invention.

  The job management system according to the present exemplary embodiment includes a plurality of distribution destination machines 101A to 101D, and the distribution destination machines 101A to 101D are connected to each other via a network 104. FIG. 1 illustrates four distribution destination machines, but as shown in FIGS. 4 and 6, more distribution destination machines constitute a job management system.

  In each of the distribution destination machines 101A to 101D, autonomous distributed job management programs 102A to 102D are installed, and network information 103A to 103D is stored.

  The network information 103A to 103D includes basic information such as MAC address, IP address, subnet mask, node name, node description, detection date and time, node type, ARP information, and FDB information, and further includes VLAN information and trunk information. Spanning tree information may be included. The ARP information is information in which the IP address is associated with the MAC address, and is included in the network information 103A to 103D when network device configuration information is requested by the SNMP protocol. The FDB information (filtering database information) is information that combines a MAC address and a port number learned when a network device communicates. The VLAN information is information on a virtual network configuration, and is a host name or a VLAN ID. The trunk information is connection interface information. Spanning tree information is information of a layer 2 protocol that operates on physical bridges and switches.

  In addition, the distribution destination machines 101A to 101D have the autonomous distributed job management programs 102A to 102D installed. However, the distribution destination machines 101A to 101D have an autonomous distributed job management program that functions as an agent by adding a new computer or changing a computer. No destination machine is envisioned. In this case, the definition file can be distributed by the method of the second embodiment described later.

  FIG. 2 is a configuration diagram of the autonomous distributed job management programs 102A to 102D.

  The autonomous distributed job management programs 102 </ b> A to 102 </ b> D include a job execution control unit 206 and a cooperation unit 210.

  The job execution control unit 206 includes a definition reading unit 207, a schedule calculation unit 208, and a job execution processing unit 209. The definition reading unit 207 reads the definition file 202 input from the input device 201. The schedule calculation unit 208 calculates a schedule from the read definition file 202. The job execution processing unit 209 executes the job according to the calculated schedule and outputs the result to the execution result log 214.

  The cooperation unit 210 includes a destination list reading unit 211, a transmission destination setting unit 212, and an event communication unit 213. The destination list reading unit 211 reads the destination list 215 and sets the destination in the destination setting unit 212. The event communication unit 213 transmits / receives an event to / from the destination set in the destination setting unit 212.

  The definition file 202 includes job definition information 203, job schedule definition information 204, and business flow information 205.

  The job definition information 203 includes items such as the contents of the job to be executed, the job execution schedule (day and time), the priority, the presence / absence of a substitute job, etc., and these items are changed depending on the contents of the job. Not limited to items.

  The job schedule definition information 204 includes calendar information indicating the entire job schedule, the name of the business flow to be executed, the execution start date, the execution time, and the execution cycle. These items are not essential for each job. , Changed by job attributes. The calendar information varies depending on the purpose of execution of the job, such as calendar for general use, in-house use, project use, and job execution machine.

  The business flow information 205 includes a business flow name, a name of a job to be executed, a command, and parameter (job dependency) information. The parameter stores information on the job connection destination and the job of the dependency destination, and the time priority at the time of job connection may be set. As for the temporal priority at the time of job connection, when the preceding job is delayed and the time for executing the succeeding job has passed, it is necessary to control the execution of the job in consideration of the priority. There are also jobs that do not need to be executed. If there is no job to be referenced, the parameter is set to a null value and the execution of the job is stopped.

  FIG. 3 is a block diagram showing a physical configuration of the distribution destination machine 101A. In FIG. 3, the distribution destination machine 101A will be described, but the other distribution destination machines 101B to 101D have the same configuration.

  The distribution destination machine 101 </ b> A of this embodiment is configured by a computer having a processor (CPU) 1, a memory 2, an auxiliary storage device 3, and a communication interface 4.

  The processor 1 executes a program stored in the memory 2. The memory 2 includes a ROM that is a nonvolatile storage element and a RAM that is a volatile storage element. The ROM stores an immutable program (for example, BIOS). The RAM is a high-speed and volatile storage element such as a DRAM (Dynamic Random Access Memory), and temporarily stores a program executed by the processor 1 and data used when the program is executed.

  The auxiliary storage device 3 is a large-capacity non-volatile storage device such as a magnetic storage device (HDD) or a flash memory (SSD), and stores a program executed by the processor 1 and data used when the program is executed. To do. That is, the program is read from the auxiliary storage device 3, loaded into the memory 2, and executed by the processor 1.

  The communication interface 4 is a network interface device that controls communication with other devices according to a predetermined protocol.

  The distribution destination machine 101A may include an input interface 5 and an output interface 8. The input interface 5 is an interface that receives input from an operator, and is a mouse, a keyboard, a touch panel, or the like. The output interface 8 is connected to a display device 9, a printer, and the like, and is an interface that outputs the execution result of the program in a form that can be visually recognized by the operator.

  The program executed by the processor 1 is provided to the distribution destination machine 101A via a removable medium (CD-ROM, flash memory, etc.) or a network, and is stored in the nonvolatile auxiliary storage device 3 that is a non-temporary storage medium. . For this reason, the behavior visualization apparatus 101 may have an interface for reading data from a removable medium.

  The distribution destination machine 101A is a computer system that is configured on a single computer or a plurality of computers that are logically or physically configured, and operates on separate threads on the same computer. It may be possible to operate on a virtual machine constructed on a plurality of physical computer resources. Each functional unit of the distribution destination machine 101A may be realized on a different computer.

  FIG. 4 is a diagram for explaining creation of the destination list 305.

  The job management system is configured to be divided into a plurality of subnets 301A to 301E on the network. In each of the subnets 301A to 301E, leader machines 303A to 303E and machines 302A to 302E other than the leader machine are installed. The machines 302A to 302E other than the leader machine are grouped.

  In this embodiment, the reader list 304 is created before the destination list 305 is created. The leader list 304 is a list of IP addresses of the leader machines 303A to 303E of the subnets 301A to 301E. The leader list 304 may be created in advance by an administrator.

  Upon receiving the reader list 304, one reader machine 303C deletes its own IP address from the reader list 304 and writes its own IP address in the destination list 305. Next, the reader machine 303C multicasts a request for an IP address using the UDP protocol, acquires the IP address of another machine 302C in the subnet 301C, and writes the acquired IP address in the destination list 305. Then, the leader machine 303C transmits the leader list 304 and the destination list 305 to the next leader machine.

  FIG. 5 is a flowchart of processing for creating the destination list 305.

  Upon receiving the reader list 304 and the destination list 305, the leader machine 303C repeats the processing of steps 402 to 409 until it finds its own IP address and a valid IP address to be the next destination (step 401). In the loop, the address is read from the leader list 304 until the end of the leader list 304 is reached (step 402).

  Then, it is determined whether the read IP address matches its own IP address (step 403). As a result, when the read IP address matches the own IP address, the own IP address is written in the destination list 305 (step 407), and the own IP address is deleted from the reader list 304 (step 408).

  On the other hand, if the IP address read in step 403 is not its own IP address, a Ping command is transmitted to the read IP address to determine whether the IP address is a response (step 404). If there is no response to the Ping command, it is determined that the IP address is not valid, and is deleted from the reader list (step 406). On the other hand, if there is a response to the Ping command, it is determined that the IP address is a valid IP address, and the IP address is set in the next destination reader machine (step 405). In the present embodiment, whether the machine is alive or not is confirmed by the Ping command, but the active or inactive state of the machine may be confirmed by another health check method.

  After the own IP address and the IP address of the next destination are found, the loop is exited, and a request for the IP address is issued by UDP to the machines in the subnet (409). Upon receiving the request UDP request, the distribution destination machine 302C returns the IP address of the own machine to the reader machine 303C.

  The reader machine 303C writes the received IP address of the distribution destination machine 302C in the destination list 305 (step 410).

  The leader machine 303C sends the leader list 304 and the destination list 305 to the next destination leader machine determined in step 405 (step 410).

  The plurality of reader machines 303A to 303E execute the processing shown in FIG. 5 in order, and finally all addresses are deleted from the reader list 304. The reader machine that has received the reader list lastly holds the destination list 305 in which all the distribution destination machines are written, and distributes the destination list 305 to all the machines written in the destination list 305.

  FIG. 6 is a diagram for explaining the distribution of the definition file 502.

  One of the plurality of distribution destination machines is defined as the definition machine 503, and the user uses the definition machine 503 to create a definition file to be distributed to the distribution destination machine.

  Next, the distribution destination machine is divided into a plurality of groups. For this reason, the definition machine 503 counts the number of IP addresses written in the destination list 305 and acquires the total number of machines. The definition machine 503 divides the IP address (distribution destination machine) of the destination list 305 by the division number N determined according to the total number of machines, and determines the division number of the distribution tree to which the definition file 505 is distributed. The division number N may be determined according to the number of layers and the number of nodes. The division number N may be determined in consideration of the load on each distribution destination machine. The number of divisions N may be determined by the user.

  In the case shown in FIG. 6, the division is performed with N = 2. However, for example, when distributing the load of each distribution destination machine at the time of distributing the definition file, the division number N = 2, the number of distribution clients is large, and the file distribution is performed. In order to shorten the time, the number of divisions N should be increased (for example, N = 10).

  In addition, when the number of lower nodes in a hierarchy exceeds a predetermined threshold, the number of divisions N may be increased in order to increase the number of machines from which definition files are distributed and execute parallel processing to increase the speed. Since the expansion of the hierarchy is proportional to the number of nodes that can be distributed, the number of divisions can be increased when the hierarchy increases.

  More specifically, in the example illustrated in FIG. 6, the 30 destinations described in the destination list 501 are divided by 2, and the destination list 501A and the destination list 501B are created.

  The definition machine 503 includes a destination list 504 in which the IP address (192.168.3.8) is deleted from the distribution destination machine 506 having the last IP address “192.168.3.8” of the destination list 501A and the definition. The destination list 505 in which the file 502 is sent and the IP address (192.168.2.3) is deleted to the distribution destination machine 507 having the last IP address “192.168.2.3” in the destination list 501B A definition file 502 is sent.

  Similarly, upon receiving the destination list 504 and the definition file 502, the distribution destination machine 506 divides the destination list 504 by an arbitrary number N (2 in the example shown in FIG. 6), and the last IP of the divided destination list. The destination lists 508 and 509 from which the last IP address has been deleted and the definition file 502 are distributed to machines having addresses. Similarly, upon receiving the destination list 505 and the definition file 502, the distribution destination machine 507 divides the destination list 505 by an arbitrary number N (2 in the example shown in FIG. 6), and the last IP of the divided destination list. The destination lists 510 and 511 from which the last IP address is deleted and the definition file 502 are distributed to machines having addresses.

  The above-described procedure is repeated until all IP addresses in the destination list are deleted, and the definition file is distributed to the distribution destination machine.

  As described above, according to the first embodiment of the present invention, one reader machine 303C collects addresses of machines to which the definition file 502 is distributed within a predetermined range (subnet) of the network 104. Since the definition file 502 is distributed to the machines with the collected addresses, it is possible to dynamically cope with changes in the system configuration such as addition / deletion of machines, and the administrator does not have to manage and register the distribution destination machines. be able to.

  Further, one reader machine 303C requests an address from a machine in the subnet of the network 104, collects addresses of computers in the subnet, and stores the collected address and the address of the reader machine 303C in the destination list 305. Since it is recorded and sent to a computer that collects addresses in another subnet, a destination list in which distribution destinations in the subnet are recorded can be created even when the administrator cannot manage or register the distribution destination machines of the definition file. In addition, since addresses are collected from the job management program installed on the distribution destination machine on the network, the distribution destination machine on which the job management program is installed can be automatically identified and the file can be distributed automatically.

  One reader machine 303C maintains a reader list 304 in which addresses of reader machines that collect addresses of distribution destination machines of definition files are recorded, and a machine at an address selected from the reader list 304 is in an active state. The address of the computer determined to be inactive is deleted from the reader list 304, the computer determined to be active is set as the destination of the reader list 304, and set Since the leader list 304 and the destination list 305 are sent to the destination, if the administrator creates the leader list 304, a destination list in which distribution destinations in the subnet are recorded can be created.

  Further, each distribution destination machine determines the number N of distribution trees to distribute the definition file 502 in consideration of the number of computers constituting the job management system and the load status of the distribution destination machine. Since each machine determines the distribution destination and distributes the definition file, it is possible to avoid concentration of load on a specific machine. In addition, the number of divisions can be determined based on the balance between distribution speed and machine load.

<Example 2>
Next, a second embodiment of the present invention will be described.

  In the first embodiment described above, the destination list is created by collecting the IP addresses of the subordinate distribution destination machines using the leader list. However, in the second embodiment, the machine dependency is acquired from the job definition. Determine the distribution destination of the definition file.

  FIG. 7 is a flowchart of processing for distributing a definition file to a destination generated from the machine dependency acquired from the job definition.

  The job definition includes job dependency as a parameter.

  When the machine dependency is acquired from the job definition and the distribution destination is determined (step 601), the IP address of the job management program at the connection destination is acquired from the dependency recorded as a parameter in the job definition (step 601). 602), Ping is transmitted to the acquired IP address, and whether the machine is alive or not is confirmed (step 603). If there is no Ping reply within a predetermined time, it is determined that the autonomous distributed job management program is not installed in the machine with the IP address or the machine with the IP address is inactive, and the upper-level job management A no return reply is transmitted to the program (step 607), and the distribution of the definition file is terminated (step 608).

  On the other hand, if there is a Ping response and the machine with the IP address that sent the Ping is alive, the definition file 502 is sent to the connection destination job management program (step 604).

  The connection destination job management program further acquires the IP address of its connection destination job management program from the dependency recorded as a parameter in the job definition, sends Ping to the acquired IP address, and the machine After confirming that it is alive, the definition file 502 is sent to the connection-destination job management program (step 605).

  When the distribution of the definition file 502 is completed to the lower-level job management program, the job management program sends the distribution end information to the higher-level job management program (step 606).

  By repeating the processing in step 606 in each hierarchy of the tree structure, the upper hierarchy job records information such as the IP address of the job management program for which the distribution of the definition file 502 has been completed, and stores it in the upper job management program. Send it.

  A job management program whose connection destination job information is not described in the parameter sends information indicating that there is no distribution destination to a higher-level job management program. The leader machine can confirm that the definition file 502 has been distributed to all the machines to which the information is to be distributed at the stage when the information indicating that there is no distribution destination or the distribution end is received from all the machines. Can have a destination list containing information on the IP addresses of When the leader machine next distributes the definition file 502, the reader machine may distribute the definition file 502 based on the destination list.

  FIG. 8 is a flowchart of processing for finding a machine having a job management program. The process shown in FIG. 8 is executed when the leader machine does not have a destination list and the transmission destination and transmission range are unknown.

  First, pings are transmitted all at once from the leader machine (step 701). Ping is transmitted to all ranges specified by the subnet address. Note that if a Ping is transmitted beyond the subnet address range, a machine unrelated to the distribution of the definition file may respond to the Ping, and information may be leaked, so the transmission range is limited.

  The reply of Ping confirms whether the machine is alive or not (Step 702), and transmits an SNMP command including an identifier (OID: Object Identifier) that can uniquely identify management information to the machine that has returned Ping (Step 703). ).

  The distribution destination machine returns a GetResponse command (step 704).

  Further, the distribution destination machine determines whether the machine that transmitted the SNMP command is the leader machine that manages itself based on the identifier of the management information included in the received SNMP command (step 705). If the machine that transmitted the SNMP command is a leader machine, the network device information is transmitted (step 706).

  Thereafter, the leader machine extracts the IP address from the network device information, and distributes the definition file 502 to the machine with the extracted IP address (step 707). When the distribution of the definition file 502 is completed, the distribution completion information is sent to the upper-level job management program (step 708).

  The distribution destination machine receiving the information distribution distributes the distribution information to the machine scheduled to be connected to the own machine from the dependency information of its own job management program. By including a terminal that includes an address and has returned a GetResponse command in the destination list, information can be distributed without having an IP address list in advance.

  Even if the leader machine is out of order, data can be sent to a machine that needs information to be sent if the machine is in the same network as the subnet address.

  As described above, according to the second embodiment of the present invention, one reader machine analyzes the dependency relationship of the computers included in the definition file 502 to be distributed and becomes the distribution destination of the definition file 502. Since the machine address is collected, even if the administrator cannot manage or register the distribution destination machine of the definition file, the address of the distribution destination can be collected, saving the administrator from managing and registering the distribution destination machine. Can do.

  In addition, one leader machine determines whether the machine at the collected address is active, distributes the data to the machine determined to be active, and if there is no active machine, Since it is determined that there is no distribution destination computer, it is possible to reduce the risk of distributing the definition file to an unauthorized machine due to impersonation.

  The present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the scope of the appended claims. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and the present invention is not necessarily limited to those having all the configurations described. A part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Moreover, you may add the structure of another Example to the structure of a certain Example. In addition, for a part of the configuration of each embodiment, another configuration may be added, deleted, or replaced.

  In addition, each of the above-described configurations, functions, processing units, processing means, etc. may be realized in hardware by designing a part or all of them, for example, with an integrated circuit, and the processor realizes each function. It may be realized by software by interpreting and executing the program to be executed.

  Information such as programs, tables, and files that realize each function can be stored in a storage device such as a memory, a hard disk, and an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, and a DVD.

  Further, the control lines and the information lines are those that are considered necessary for the explanation, and not all the control lines and the information lines that are necessary for the mounting are shown. In practice, it can be considered that almost all the components are connected to each other.

101A to 101D: Distribution destination machines 102A to 102D: Autonomous distributed job management programs 103A to 103D: Network information 104: Network 201: Input device 202: Definition file 203: Job definition information 204: Job schedule definition information 205: Business flow information 206: Job execution control unit 207: Definition reading unit 208: Schedule calculation unit 209: Job execution processing unit 210: Cooperation unit 211: Destination list reading unit 212: Delivery destination setting unit 213: Event communication unit 214: Execution result log 215: Destination lists 301A to 301E: Subnets 302A to 302E: Machines 303A to 303E other than the leader machine: Reader machine 304: Reader list 305: Destination lists 401 to 410: Flowcharts 501, 501A, 501 , 504,505,508～511,516～523: destination list 502: Definition File 503: Definition machine 506,507,512～515,524～547: distribution destination machine

Claims (12)

A computer system that distributes data for managing jobs,
The computer system is configured by connecting a plurality of computers that execute jobs via a network,
Each of the plurality of computers has a processor that executes a program, and a storage device that stores the program,
A computer system in which one of the plurality of computers collects addresses of computers to which the data is distributed within a predetermined range of the network, and distributes the data to the computers having the collected addresses. The computer system according to claim 1,
One of the plurality of computers is
Request addresses from a computer in the subnet of the network, collect the addresses of the computers in the subnet,
Record the collected address and the address of the computer in the destination list,
A computer system that sends the destination list to a computer that collects addresses in another subnet. The computer system according to claim 2,
One of the plurality of computers is
Keep a reader list with the addresses of the computers that collect the addresses of the computers to which data is distributed,
Determining whether the calculator at the address selected from the reader list is active;
Delete the address of the computer determined to be inactive from the reader list,
Set the computer that is determined to be active to the destination of the reader list,
A computer system for sending the reader list and the destination list to the set destination. The computer system according to claim 1,
Each of the plurality of computers is a computer system that determines the number of distribution trees to distribute the data in consideration of the number of computers constituting the computer system and the load status of the computers. The computer system according to claim 1,
One of the plurality of computers is a computer system that analyzes a dependency relationship of computers included in the data to be distributed and collects addresses of computers to which the data is distributed. The computer system according to claim 5,
One of the plurality of computers is
Determining whether the collected address calculator is active;
Distribute the data to computers determined to be active,
A computer system that determines that there is no distribution destination of the data when there is no computer in an active state. A data distribution method in a job management system,
The job management system is configured by connecting a plurality of computers that execute jobs via a network,
Each of the plurality of computers has a processor that executes a program, and a storage device that stores the program,
A data distribution method in which one of the plurality of computers collects addresses of computers to which the data is distributed within a predetermined range of the network, and distributes the data to the computers having the collected addresses. The data distribution method according to claim 7,
One of the plurality of computers is
Requesting an address from a computer in the subnet of the network, collecting the address in a computer in the subnet,
Record the collected address and the address of the computer in the destination list,
A data distribution method for sending the destination list to a computer that collects addresses in another subnet. The data distribution method according to claim 8, comprising:
One of the plurality of computers is
Keep a reader list with the addresses of the computers that collect the addresses of the computers to which data is distributed,
Determining whether the calculator at the address selected from the reader list is active;
Delete the address of the computer determined to be inactive from the reader list,
Set the computer that is determined to be active to the destination of the reader list,
A data distribution method for sending the reader list and the destination list to the set destination. The data distribution method according to claim 7,
A data distribution method in which each of the plurality of computers determines the number of distribution trees to distribute the data in consideration of the number of computers constituting the job management system and the load status of the computers. The data distribution method according to claim 7,
A data distribution method in which one of the plurality of computers analyzes a dependency relationship of computers included in the data to be distributed and collects addresses of computers to which the data is distributed. A data distribution method according to claim 11, comprising:
One of the plurality of computers is
Determining whether the collected address calculator is active;
Distribute the data to computers determined to be active,
A data distribution method for determining that there is no distribution destination of the data when there is no computer in an active state.

Images