JP6476718B2 - INFORMATION PROCESSING SYSTEM, PROGRAM, AND INFORMATION PROCESSING METHOD - Google Patents

Description

  The present invention relates to an information processing system.

  A building block system (hereinafter sometimes referred to as a BB system) that realizes a large scale system by connecting a plurality of information processing apparatuses is known. Each information processing device of the BB system includes a processor (processing device) and a memory, and can process information. The information processing apparatus may be a single computer in which a processor and a memory are housed in a single housing, or may be a single system board on which a processor and a memory are mounted. Each information processing apparatus has a communication function, and can process information in cooperation with other information processing apparatuses.

  Each information processing apparatus in a certain BB system includes a service processor (SP) that manages hardware devices such as a processor (processing apparatus) and a memory in the own information processing apparatus. The SP includes, for example, a MPU (Micro-Processing Unit), a non-volatile memory such as a flash memory, and a volatile memory. Each SP manages hardware devices such as a processor (processing device) and a memory in its own information processing device, and stores device information of the hardware device as “SP data”.

  The BB system as described above is an example of an information processing system including a plurality of processing devices and a plurality of management devices that manage the plurality of processing devices. That is, each SP in the BB system is a specific example of a management device, and manages one or more processing devices.

  In some BB systems, SPs are classified into a master SP, a standby SP that is a backup of the master SP, and a slave SP. An SP in any one information processing apparatus among a plurality of information processing apparatuses operates as a master SP. The master SP is an SP that supervises the standby SP and the slave SP, and collects SP data on various hardware devices managed by the standby SP and slave SP from the standby SP and slave SP. Since the master SP is an SP that supervises the standby SP and the slave SP, hardware devices managed by the standby SP and slave SP can be managed via the standby SP and slave SP.

  The master SP also manages hardware devices such as a processor (processing device) and a memory in the information processing apparatus. Among the SP data collected by the master SP, the standby SP receives at least SP data of hardware devices managed by the slave SP from the master SP, and when a failure occurs in the master SP, the slave SP substitutes for the slave SP. It is an SP that controls the SP. The standby SP manages hardware devices such as a processor (processing device) and a memory in the own information processing device. The slave SP manages hardware devices such as a processor (processing device) and a memory in its own information processing device.

  A multiplexed service processor has also been proposed that includes a time synchronization processing unit that synchronizes time between service processors, an interprocessor monitoring processing unit that monitors interprocessor communication, and a diagnosis processing unit that operates with the synchronized time as a trigger. (See, for example, Patent Document 1). The diagnosis processing unit autonomously sequentially executes a series of self-diagnosis processing of writing processing of diagnosis data to the shared storage area in the unit, reading processing from the shared storage area, and monitoring processing of interprocessor communication. The diagnostic processing unit also performs switching processing to a standby service processor.

JP, 2011-65521, A

  In the information processing system having a plurality of management devices, when a failure occurs in the communication path used to transmit data from the first management device to the second management device, the data can not be transmitted.

  In one aspect, the present invention is directed to enabling transmission and reception of data between management devices to be continued even when a failure occurs in a communication path between management devices.

An information processing system according to an aspect includes a plurality of processing devices and a plurality of management devices that manage the plurality of processing devices. The first management device stores data. The second management device receives data from the first management device via the communication path, and stores the received data. When a failure occurs in the communication path, the first management device selects a first management device and a second management device from among the one or more other management devices based on the state of the other one or more management devices. Select a third management device connected to the management device of Then, the first management device transmits data to the second management device via the third management device. The second management device transmits the second data stored by the second management device to the first management device via the communication path or another communication path. If a failure occurs in the communication path used by the second management device for transmission of the second data or the other communication path, the second management device is based on the status of the one or more other management devices. A fourth management apparatus different from the third management apparatus or the third management apparatus is selected from the one or more other management apparatuses, and the selected third management apparatus or the selected fourth management apparatus The second data is transmitted to the first management device via the management device. The state of each of the other one or more management devices is an indicator based on the degree to which the state of one or more processing devices managed by the other management device adversely affects data transmission via the other management device. Represented by a value. The first management device selects a third management device based on the index value. The second management device selects the third management device or the fourth management device based on the index value.

  Even if a failure occurs in the communication path between the management devices, data transmission and reception can be continued between the management devices.

It is a figure explaining the example of the BB system concerning this embodiment. It is a figure explaining an example (the 1) of processing of BB system when fault occurs in a channel. It is a figure explaining the example (the 1) of the hardware constitutions of an information processor. It is a figure explaining the example (the 2) of the hardware constitutions of an information processor. It is a figure explaining the example (the 2) of processing of BB system when a failure occurs in a communication way. It is a figure explaining the example (the 3) of processing of BB system when a failure occurs in a communication way. It is a figure explaining an example of processing which forms an alternative course. It is a figure explaining the example of passage information. It is a figure explaining the example of coefficient information. It is a flowchart explaining the example of the alternative path formation process of master SP. It is a flowchart explaining the example of the alternative path formation process of standby SP. It is a flowchart explaining the example of calculation processing of the risk index value of slave SP. It is a flowchart explaining the example of calculation processing of the risk index value of slave SP. It is a flowchart explaining the example of selection processing of slave SP. It is a flowchart explaining the example of selection processing of slave SP. It is a flowchart explaining the example (the 1) of formation processing of an alternative path. It is a flowchart explaining the example (the 2) of formation processing of an alternative path. It is a flowchart explaining the example (the 3) of formation processing of an alternative path. It is a flowchart explaining the example of the processing after the failure of the communication path is recovered. It is a flow chart explaining an example of processing of recalculation of a risk index value.

Hereinafter, embodiments will be described in detail with reference to the drawings.
FIG. 1 is a diagram for explaining an example of a BB system according to the present embodiment. The BB system 100 of FIG. 1 includes an information processing apparatus 101 having a master SP 120, an information processing apparatus 102 having a standby SP 130, an information processing apparatus 103a having a slave SP 140a, and an information processing apparatus 103b having a slave SP 140b. Each information processing apparatus further includes a processor (that is, a processing device) and a memory in addition to the SP, and can process information. The information processing apparatus may be a single computer in which a processor and a memory are housed in a single housing, or may be a single system board on which a processor and a memory are mounted.

  The BB system 100 is an example of an information processing system including a plurality of processing devices and a plurality of management devices that manage the plurality of processing devices. The number of processing devices managed by each management device may be one or more arbitrary number. Each SP in FIG. 1 is an example of a management device that manages hardware devices such as a processor (processing device) and memory in the own information processing device (that is, in the information processing device in which the SP itself is installed). is there. Each SP includes, for example, an MPU, a non-volatile memory such as a flash memory, and a volatile memory.

  The information processing device 103 b is connected to an input / output device 110 (hereinafter, abbreviated as “input / output (I / O) device”). The I / O device 110 is, for example, a device such as a keyboard, a mouse, and a monitor. The I / O device 110 can be connected to any of the information processing devices 101, 102, and 103a, and may be connected to an information processing device other than the information processing device 103b. The I / O device 110 may be omitted.

  Hereinafter, the information processing apparatus 103a and the information processing apparatus 103b will be collectively referred to as "the information processing apparatus 103". Also, the slave SP 140 a and the slave SP 140 b are collectively referred to as “slave SP 140”.

  Each SP such as the master SP 120, the standby SP 130, and the slave SP 140 controls and monitors hardware devices such as a processor (processing device) and a memory in its own information processing device. Store as data. The device information of each hardware device indicates at least the state of the hardware device. Each SP also monitors the I / O device 110 when the I / O device 110 is connected to the information processing device in which the SP is installed. When the SP monitors the I / O device 110, device information of the I / O device 110 is also included in the SP data. Furthermore, each SP detects another information processing apparatus connected to the own information processing apparatus, and stores information indicating which information processing apparatus is connected to the own information processing apparatus as SP data.

  The master SP 120 is an SP that controls the standby SP 130 and the slave SP 140. In other words, the master SP 120 can manage hardware devices such as processing devices managed by the standby SP 130 and the slave SP 140 via the standby SP 130 and the slave SP 140. For the management, the master SP 120 collects SP data stored by the standby SP 130 and slave SP 140.

  The standby SP 130 is an SP that operates as a backup of the master SP 120 when the master SP 120 fails. Therefore, the standby SP 130 receives, from the master SP 120, the SP data collected by the master SP 120 (at least the SP data collected from the slave SP 130 among them) and the SP data of the master SP 120. By storing the same data as the SP data stored in the master SP 120, the standby SP 130 can be an SP that supervises the slave SP 140 as a substitute for the master SP 120 even when a failure occurs in the master SP 120.

  The information processing apparatus 101 is connected to the information processing apparatus 102 via two communication cables (ie, the communication path 151 and the communication path 152). The communication path 152 is used to transmit SP data from the master SP 120 of the information processing apparatus 101 to the standby SP 130 of the information processing apparatus 102. The communication path 151 is used to transmit SP data from the standby SP 130 of the information processing apparatus 102 to the master SP 120 of the information processing apparatus 101.

  The information processing apparatus 101 is connected to the information processing apparatus 103 a by the communication path 153 a and to the information processing apparatus 103 b by the communication path 153 b. Further, the information processing apparatus 102 is connected to the information processing apparatus 103 a by the communication path 154 a and to the information processing apparatus 103 b by the communication path 154 b.

  Although the information processing apparatus 101 and the information processing apparatus 102 are connected by two communication cables in the example of FIG. 1, the information processing apparatus 101 may be connected by one communication cable capable of bidirectional communication depending on the embodiment. And the information processing apparatus 102 may be connected.

  Here, when a failure occurs in the communication path 152, the master SP 120 can not transmit SP data to the standby SP 130 using the communication path 152. Then, a situation may occur in which the SP data stored by the master SP 120 and the SP data stored by the standby SP 130 are different. When the master SP 120 and the standby SP 130 store different SP data, the standby SP 130 can not operate as a backup of the master SP 120 when a failure occurs in the master SP 120.

  When a failure occurs in the communication path 152, the master SP 120 according to this embodiment forms another communication path via the information processing device 103a or the information processing device 103b, and transmits SP data to the standby SP 130. In order to form another communication path, the master SP 120 acquires a value indexed to information indicating the state of a hardware device such as a processing device managed by the slave SP 140. This value is calculated by the slave SP 140. The slave SP 140 may use coefficient information described later with reference to FIG. 9 when indexing information indicating the state of a hardware device such as a processing device managed by the slave SP 140. In the BB system 100 of FIG. 1, the master SP 120 selects the slave SP 140a based on the values acquired from the slave SPs 140a to 140b. The master SP 120 transmits SP data to the standby SP 130 via the slave SP 140 a using the communication path 153 a and the communication path 154 a.

  As in the above embodiment, even if a failure occurs in the communication path 152 for transmitting SP data from the master SP 120 to the standby SP 130, the master SP 120 can transmit SP data to the standby SP 130 via the slave SP 140. That is, the master SP 120 and the standby SP 130 can maintain the same SP data storage state. With the master SP 120 and the standby SP 130 storing the same SP data, the standby SP 130 can operate as a backup of the master SP 120.

  As described above, the BB system 100 is an example of an information processing system including a plurality of management devices that manage a plurality of processing devices. The master SP 120 is an example of a first management device that stores data (for example, SP data of the master SP 120 and SP data collected from the slave SP 140) among a plurality of management devices. The standby SP 130 is an example of a second management apparatus that receives the data from the first management apparatus via the communication path (for example, the communication path 152) among the plurality of management apparatuses, and stores the received data.

  When a failure occurs in the communication path, the first management device selects a third management device connected to the first management device and the second management device from among one or more other management devices. Do. For example, the slave SP 140a is an example of the third management device. The selection of the third management device is made based on the status of one or more other management devices. Specifically, the selection of the third management device may be performed based on information indicating the state of a hardware device such as a processing device mounted on the slave SPs 140a to 140b as described above.

  When the first management device selects the third management device, the first management device transmits data to the second management device via the third management device. As a result, even if a failure occurs in the communication path, it becomes possible to continue data transmission from the first management device to the second management device.

  As described above, when data transmission is performed for data synchronization between the master SP 120 and the standby SP 130, data synchronization is maintained by continuation of data transmission. If data synchronization is maintained, when a failure occurs in the master SP 120, failover from the master SP 120 to the standby SP 130 is appropriately performed.

  Note that, contrary to the above example, an embodiment is also possible in which the standby SP 130 corresponds to the first management device and the master SP 120 corresponds to the second management device. That is, in order to continue the transmission of data stored by the standby SP 130 to the master SP 120, the standby SP 130 selects the third management apparatus according to the failure of the communication path, and transmits the data via the third management apparatus. It may be transmitted to the master SP 120.

  FIG. 2 is a diagram for explaining an example (part 1) of processing of the BB system when a failure occurs in a communication path. In the BB system 100 of FIG. 2, the same components as in FIG. 1 are given the same reference numerals as in FIG. 1. In FIG. 2, I / O device 110 is not described for simplification. The I / O device 110 can be connected to any of the information processing devices 101 to 103.

  The BB system 100 of FIG. 2 includes information processing apparatuses 103a to 103n provided with slaves SP 140a to 140n, respectively. The illustration of the slave SPs 140a to 140n and the information processing apparatuses 103a to 103n in FIG. 2 is not intended to limit the number of information processing apparatuses 103 and slave SPs 140. Hereinafter, the information processing apparatuses 103a to 103n may be collectively referred to as the "information processing apparatus 103". Also, in the following, the slave SPs 140 a to 140 n may be collectively referred to as “slave SP 140”.

  The information processing apparatus 101 is connected to the information processing apparatuses 103a to 103n via the communication paths 153a to 153n. The information processing apparatus 102 is connected to the information processing apparatuses 103a to 103n via communication paths 154a to 154n, respectively.

  The master SP 120 includes a storage unit 121, a synchronization unit 122, a data control unit 123, and a detection unit 124. The standby SP 130 includes a storage unit 131, a synchronization unit 132, a data control unit 133, and a detection unit 134. The slaves SP 140a to 140n respectively include storage units 141a to 141n, data relay units 142a to 142n, management units 143a to 143n, detection units 144a to 144n, and synchronization units 145a to 145n. Hereinafter, the storage units 141 a to 141 n will be collectively referred to as “storage unit 141”. The data passing parts 142a to 142n are collectively referred to as a "data passing part 142". The management units 143a to 143n are collectively referred to as a "management unit 143". The detection units 144a to 144n are collectively referred to as "detection unit 144". The synchronization units 145a to 145n are collectively referred to as "synchronization unit 145".

  The management unit 143 of the slave SP 140 manages hardware devices such as a processor (processing device) and a memory in the own information processing device. Specifically, the management unit 143 recognizes the hardware device installed in the self-information processing device 103, and monitors a failure of the hardware device. The management unit 143 stores information of the recognized hardware device, information indicating start and stop of the partition (which will be described in detail later), and the like in the storage unit 141 as slave SP data 163. The coefficient information 164 (to be described in detail with reference to FIG. 9) is set in the storage unit 141 in advance. In FIG. 2, as an example of slave SP data 163 and coefficient information 164, slave SP data 163a and coefficient information 164a stored in the storage unit 141a in the slave SP 140 are illustrated.

  The detection unit 144 of the slave SP 140 detects a device such as the other information processing devices 101 to 102 or the I / O device 110 connected to the self information processing device 103, and stores information on the detected device as slave SP data 163. It is stored in the unit 141. The detection unit 144 further detects a failure when a failure occurs in the communication path 153 connecting the self information processing apparatus 103 and the information processing apparatus 101. Further, when a failure occurs in the communication path 154 connecting the self information processing apparatus 103 and the information processing apparatus 102, the detection unit 144 detects the failure.

  The synchronization unit 145 transmits the slave SP data 163 to the synchronization unit 122 of the master SP 120 periodically or when the slave SP data 163 of the slave SP 140 is updated. The synchronization unit 122 stores the received slave SP data 163 of the slave SP 140 in the storage unit 121. The data passing unit 142 will be described later.

  The detection unit 134 of the standby SP 130 detects devices such as the other information processing devices 101 and 103 and the I / O device 110 connected to the self information processing device 102, and stores information on the detected devices as the standby SP data 162. It memorizes in 131. The detection unit 134 stores the detected information of the information processing apparatus 103 in the storage unit 131 as route information 166. The via information 166 is a list of information processing apparatuses as candidates for selection when the standby SP 130 forms another communication path via any of the information processing apparatuses 103 a to 103 n when a failure occurs in the communication path 151. Including.

  The detection unit 134 further detects a failure when a failure occurs in the communication channel 151 and the communication channel 152. When a failure occurs in the communication path 154 connecting the self information processing apparatus 102 and the information processing apparatus 103, the detection unit 134 detects the failure.

  The synchronization unit 132 transmits the standby SP data 162 to the synchronization unit 122 of the master SP 120 periodically or when the standby SP data 162 is updated. The synchronization unit 122 of the master SP 120 stores the received standby SP data 162 in the storage unit 121. The data control unit 133 will be described later.

  The detection unit 124 of the master SP 120 detects devices such as the other information processing devices 102 to 103 and the I / O device 110 connected to the self information processing device 101, and stores information on the detected devices as master SP data 161. It memorizes in 121. The detection unit 124 stores the detected information of the information processing apparatus 103 in the storage unit 121 as route information 165. The via information 165 is a list information of the information processing apparatus 103 which is a selection candidate when a failure occurs in the communication path 152 and the master SP 120 forms another communication path via any of the information processing apparatuses 103a to 103n. including.

  Furthermore, when a failure occurs in the communication path 151 and the communication path 152, the detection unit 124 detects the failure. Further, when a failure occurs in the communication path 153 connecting the self information processing apparatus 101 and the information processing apparatus 103, the detection unit 124 detects the failure.

  The synchronization unit 122 transmits the master SP data 161 to the synchronization unit 132 of the standby SP 130 periodically or when the master SP data 161 is updated. The synchronization unit 132 of the standby SP 130 stores the received master SP data 161 in the storage unit 131. The data control unit 123 will be described later.

  The synchronization unit 122 causes the standby SP 130 to store the same SP data as the storage unit 121 of the master SP 120 so that the standby SP 130 can operate as a substitute for the master SP 120 when the master SP 120 fails. Specifically, synchronization unit 122 transmits master SP data 161 and slave SP data 163a to 163n received from slave SPs 140a to 140n to synchronization unit 132 of standby SP 130. The synchronization unit 132 of the standby SP 130 stores the master SP data 161 received from the synchronization unit 122 and slave SP data 163 a to 163 n in the storage unit 131. By storing the latest slave SP data 163a to 163n used by the master SP 120 in the storage unit 131, the standby SP 130 can manage the slave SPs 140a to 140n instead of the master SP 120 even if the master SP 120 fails. While the master SP 120 is broken, the synchronization unit 145 of the slave SP 140 transmits slave SP data 163 to the synchronization unit 132 of the standby SP 130.

  Here, when a failure occurs in the communication path 152 for transmitting SP data from the synchronization unit 122 of the master SP 120 to the synchronization unit 132 of the standby SP 130, the SP data from the master SP 120 to the standby SP 130 using the communication path 152 You can not send. Then, a situation may occur in which the SP data stored by the master SP 120 and the SP data stored by the standby SP 130 are different. When the master SP 120 and the standby SP 130 store different SP data, the standby SP 130 can not operate as a backup of the master SP 120 when a failure occurs in the master SP 120.

  Therefore, when a failure occurs in the communication path 152, the master SP 120 according to the present embodiment passes through any of the information processing apparatuses 103a to 103n (more specifically, through any of the slaves 140a to 140n). Form another communication path. This other communication path is hereinafter referred to as an alternative path. In the process of forming an alternative path, first, when the detecting unit 124 of the master SP 120 detects a failure in the communication path 152 for transmitting SP data from the master SP 120 to the standby SP 130, the data control unit Notify 123 The data control unit 123 transmits the calculation instruction to the data transit unit 142 of the slave SP 140 that can communicate with the master SP 120 included in the transit information 165 stored in the storage unit 121. The calculation instruction is an instruction for obtaining the slave SP data 163, which is information indicating the state of a hardware device such as a processing device managed by the slave SP 140, using the coefficient information 164. The data passing unit 142 that has received the calculation instruction calculates a risk index value that indexes a factor that adversely affects data transfer from the slave SP data 163 and the information included in the coefficient information 164 of the storage unit 141. Examples of factors that adversely affect data transfer are a failure of each hardware device in the information processing apparatus 103 in which the data passing unit 142 is installed, a load of each processing apparatus in the information processing apparatus 103, and the like. The data relay unit 142 transmits the calculated risk index value to the data control unit 123 of the master SP 120.

  When the data control unit 123 of the master SP 120 receives the risk index value from all slaves SP 140 that have transmitted the calculation instruction, the data control unit 123 updates the via information 165 of the storage unit 121 based on the risk index value. The data control unit 123 of the master SP 120 selects one slave SP 140 based on the risk index value (to be described later in detail with reference to FIG. 9). The data control unit 123 of the master SP 120 forms an alternative path via the selected slave SP 140.

  The risk index value is an example of an index value based on the degree to which the state of hardware devices such as one or more processing devices managed by the slave SP 140 adversely affects data transmission via the slave SP 140. In the present embodiment, the state of the slave SP 140 is represented by the risk index value. The state represented by the risk index value is, in other words, the degree to which the slave 140 is suitable as a relay point through which the alternative route passes. In the present embodiment, one slave SP 140 is selected based on the condition represented by the risk index value.

  When the alternative path is formed, the synchronization unit 122 of the master SP 120 outputs the SP data (master SP data 161 and slave SP data 163) to be transmitted to the standby SP 130 to the data control unit 123. The data control unit 123 transmits the SP data received from the synchronization unit 122 to the data passing unit 142 of the selected slave SP 140. The data passing unit 142 transmits the received SP data to the data control unit 133 of the standby SP 130. The data control unit 133 outputs the received SP data to the synchronization unit 132. The synchronization unit 132 stores the SP data received from the data control unit 133 in the storage unit 131.

  Thus, even if a failure occurs in the communication path 152 for transmitting SP data from the master SP 120 to the standby SP 130 by the master SP 120 forming an alternative path, the master SP 120 transmits SP data to the standby SP 130. You can keep doing. By the master SP 120 and the standby SP 130 storing the same SP data, the standby SP 130 can maintain an operable state as a backup of the master SP 120.

  FIG. 3 is a diagram for explaining an example (part 1) of the hardware configuration of the information processing apparatus. The information processing apparatuses 101 to 103 illustrated in FIGS. 1 and 2 have the same hardware configuration. Therefore, the information processing apparatuses 101 to 103 are collectively referred to as an “information processing apparatus 200”. The information processing apparatus 200 includes a CPU Memory Unit (CMU) 210 and a service processor (SP) 220. CPU is an abbreviation of Central Processing Unit.

  The CMU 210 includes a processor 211 and a memory 212. The processor 211 is an example of a processing device managed by the SP 220. The processor 211 may be any processing circuit, and may be, for example, a central processing unit (CPU). The memory 212 appropriately stores data obtained by the operation of the processor 211 and data used for the processing of the processor 211.

  Although FIG. 3 shows an example in which the information processing apparatus 200 includes one CMU 210, the information processing apparatus 200 may include two or more CMUs 210. Also, one CMU 210 may have two or more processors 211.

  The SP 220 includes a processor 221 and a memory 222. The processor 221 may be any processing circuit, and may be, for example, a central processing unit (CPU). In the master SP 120, the processor 221 operates as a detection unit 124, a synchronization unit 122, and a data control unit 123 by executing a program. In the standby SP 130, the processor 221 operates as a detection unit 134, a synchronization unit 132, and a data control unit 133 by executing a program. In the slave SP 140, the processor 221 operates as a synchronization unit 145, a detection unit 144, a data relay unit 142, and a management unit 143 by executing a program.

  In the master SP 120, the memory 222 is used as the storage unit 121, and holds a copy of SP data of the standby SP 130, a copy of SP data of the slave SP 140, and SP data of the master SP 120 and passing information 165. In the standby SP 130, the memory 222 is used as the storage unit 131, and holds the SP data of the standby SP 130, the copy of the SP data of the slave SP 140, the copy of the SP data of the master SP 120, and the routing information 165. In the slave SP 140, the memory 222 is used as the storage unit 141, and holds the SP data of the slave SP 140 and the coefficient information 164. Furthermore, the memory 222 appropriately stores data obtained by the operation of the processor 221 and data used for the processing of the processor 221.

  FIG. 4 is a diagram for explaining an example (part 2) of the hardware configuration of the information processing apparatus. The information processing apparatus 200 includes a processor 11, a memory 12, a bus 15, an external storage device 16, a network connection device 19, and an SP 20. Further, as an option, the information processing device 200 may include an input device 13, an output device 14, and a medium drive device 17. Alternatively, at least one of the input device 13, the output device 14, and the medium drive device 17 may be external to the information processing device 200 and connected to the information processing device 200. The information processing device 200 may be realized by, for example, a computer.

  The processor 11 is the processor 211 of the CMU 210 of FIG. The memory 12 is a memory 212 of the CMU 210. The processor 11 can load a program stored in the external storage device 16 into the memory 12 and execute the program, for example. The network connection device 19 is used to communicate with other devices. The network connection device 19 is, for example, a communication circuit such as a network adapter. SP20 is SP220 of FIG.

  The input device 13 may be, for example, a button, a keyboard, a mouse or the like, and the output device 14 may be a display or the like. The bus 15 connects the processor 11, the memory 12, the input device 13, the output device 14, the external storage device 16, the medium drive device 17, the network connection device 19, and the SP 20 so as to mutually exchange data. The SP 20 can communicate with another SP 20 mounted on another information processing apparatus 200 via the network connection device 19 (and a physical communication path such as a cable). The external storage device 16 stores programs and data. The external storage device 16 may be, for example, a Hard Disk Drive (HDD) or a Solid-State Drive (SSD). The information stored in the external storage device 16 is appropriately provided to the processor 11 or the like. The medium drive device 17 can transmit data stored in the memory 12 or the external storage device 16 to the portable recording medium 18, and can read a program, data, etc. from the portable recording medium 18. Here, the portable recording medium 18 is any portable recording such as a floppy disk, a Magneto-Optical (MO) disk, a Compact Disc Recordable (CD-R), a Digital Versatile Disk Recordable (DVD-R), etc. It can be a medium. The portable recording medium 18 may be a semiconductor memory card such as a flash memory, and the medium driving device 17 may be a reader and a writer for the memory card.

  FIG. 5 is a diagram for explaining an example (part 2) of the process of the BB system when a failure occurs in the communication path. In the BB system 100 of FIG. 5, the same components as in FIG. 2 are denoted by the same reference numerals as in FIG.

  When a failure occurs in the communication path 151 for transmitting SP data from the synchronization unit 132 of the standby SP 130 to the synchronization unit 122 of the master SP 120, SP data can be transmitted from the standby SP 130 to the master SP 120 using the communication path 151. It disappears. Then, a situation may occur in which the SP data stored by the master SP 120 and the SP data stored by the standby SP 130 are different. When the master SP 120 and the standby SP 130 store different SP data, the standby SP 130 can not operate as a backup of the master SP 120 when a failure occurs in the master SP 120.

  Therefore, when a failure occurs in the communication path 151, the standby SP 130 according to the present embodiment starts processing for forming an alternative path of the communication path 151 in which the failure has occurred. In the process of forming an alternative path, first, when the detecting unit 134 of the standby SP 130 detects a failure in the communication path 151 for transmitting SP data from the standby SP 130 to the master SP 120, the data control unit Notify 133 The data control unit 133 identifies the slave SP 140 that can communicate with the standby SP 130 based on the route information 166 stored in the storage unit 131. The data control unit 133 transmits, to the data passing unit 142 of the identified slave SP 140, a calculation instruction for obtaining the index of the slave SP data 163 using the coefficient information 164. The data passing unit 142 that has received the calculation instruction calculates a risk index value that indexes a factor that adversely affects data transfer from the slave SP data 163 and the coefficient information 164 of the storage unit 141. The data relay unit 142 transmits the calculated risk index value to the data control unit 133 of the standby SP 130.

  When the data control unit 133 of the standby SP 130 receives the risk index value from all slaves SP 140 that has transmitted the calculation instruction, the data control unit 133 updates the via information 166 of the storage unit 131 based on the risk index value. The data control unit 133 of the standby SP 130 selects one slave SP 140 based on the risk index value (details will be described later with reference to FIG. 9). The data control unit 133 of the standby SP 130 forms an alternative path via the selected slave SP 140.

  When the alternative path is formed, the synchronization unit 132 of the standby SP 130 outputs the SP data (standby SP data 162) to be transmitted to the master SP 120 to the data control unit 133 of the standby SP 130. The data control unit 133 of the standby SP 130 transmits the SP data received from the synchronization unit 132 to the data passing unit 142 of the selected slave SP 140. The data passing unit 142 transmits the received SP data to the data control unit 123 of the master SP 120. The data control unit 123 of the master SP 120 outputs the received SP data to the synchronization unit 122. The synchronization unit 122 stores the SP data received from the data control unit 123 in the storage unit 121.

  Thus, even if a failure occurs in the communication path 151 for transmitting SP data from the standby SP 130 to the master SP 120 by the standby SP 130 forming an alternative path, the standby SP 130 transmits SP data to the master SP 120. You can keep doing. By the master SP 120 and the standby SP 130 storing the same SP data, the standby SP 130 can maintain an operable state as a backup of the master SP 120.

  In the BB system 100 of FIGS. 2 and 5, the master SP 120 or the standby SP 130, which is the SP data transmission source, detects a failure in the communication path used for SP data transmission, and forms an alternative path. The master SP 120 and the standby SP 130 can store the same SP data as the transmission source SP forms an alternative path.

  FIG. 6 is a diagram for explaining an example (part 3) of the process of the BB system when a failure occurs in the communication path. In the BB system 100 of FIG. 6, the same components as in FIG. 2 are denoted by the same reference numerals as in FIG. In the BB system 100 of FIG. 6, failures occur in both the communication path 152 for transmitting SP data from the master SP 120 to the standby SP 130 and the communication path 151 for transmitting SP data from the standby SP 130 to the master SP 120. An example of

  When detecting a failure on the communication path 152 for transmitting SP data from the master SP 120 to the standby SP 130, the detection unit 124 of the master SP 120 notifies the data control unit 123 that the failure is detected. Similarly, when detecting a failure on the communication path 151 for transmitting SP data from the standby SP 130 to the master SP 120, the detection unit 134 of the standby SP 130 notifies the data control unit 133 that the failure is detected.

  The data control unit 123 specifies the slave SP 140 that can communicate with the master SP 120 based on the via information 165 stored in the storage unit 121. The data control unit 123 transmits, to the data passing unit 142 of the specified slave SP 140, a calculation instruction for obtaining the index of the slave SP data 163 using the coefficient information 164.

  The data control unit 133 identifies the slave SP 140 that can communicate with the standby SP 130 based on the route information 166 stored in the storage unit 131. The data control unit 133 transmits, to the data passing unit 142 of the identified slave SP 140, a calculation instruction for obtaining the index of the slave SP data 163 using the coefficient information 164. Note that the processing of the data control unit 123 and the data control unit 133 is performed in different information processing apparatuses, so the execution timing may be different.

  When receiving the calculation instruction from the data control unit 123 of the master SP 120, the data passing unit 142 calculates a risk index value, and transmits the calculated risk index value to the data control unit 123 of the master SP 120. Also, upon receiving a calculation instruction from the data control unit 133 of the standby SP 130, the data passing unit 142 calculates a risk index value, and transmits the calculated risk index value to the data control unit 133 of the standby SP 130.

  When transmitting the risk index value, the data relay unit 142 of the slave SP 140 also transmits reservation information indicating whether the slave SP 140 has already been selected as an alternative path.

  When the data control unit 123 of the master SP 120 receives the risk index value and the reservation information from all slaves SP 140 that have transmitted the calculation instruction, the data control unit 123 adds the risk index value and the reservation information to the via information 165 of the storage unit 121. The data control unit 123 of the master SP 120 selects one slave SP 140 b based on the risk index value (to be described later in detail with reference to FIG. 9). The slave SP 140, which is indicated by the reservation information that has already been selected as the alternative route, is excluded from the selection targets. The data control unit 123 of the master SP 120 forms an alternative path via the selected slave SP 140 b. The selected slave SP 140 b stores information indicating that the slave SP 140 b itself is selected as an alternative path in the storage unit 141 b as coefficient information 164 in the process of forming the alternative path. When the formation of the alternative path is completed, the data control unit 123 of the master SP 120 updates the passing information 165 of the storage unit 121 so as to indicate that the formation of the alternative path is completed.

  When the data control unit 133 of the standby SP 130 receives the risk index value and the reservation information from all slaves SP 140 that have transmitted the calculation instruction, the data control unit 133 adds the information of the risk index value and the reservation information to the via information 166 of the storage unit 131. The data control unit 133 of the standby SP 130 selects one slave SP 140a based on the risk index value (to be described later in detail with reference to FIG. 9). The slave SP 140b, which is indicated by the reservation information that has already been selected as the alternative route, is excluded from the selection targets. The data control unit 133 of the standby SP 130 forms an alternative path via the selected slave SP 140a. The selected slave SP 140 a stores information indicating that the slave SP 140 a is selected as an alternative path in the storage unit 141 a as slave SP data 163 in the process of forming the alternative path. When the formation of the alternative path is completed, the data control unit 133 of the standby SP 130 updates the passing information 165 of the storage unit 131 so as to indicate that the formation of the alternative path is completed.

  If there is only one slave SP 140 in the BB system 100, the master SP 120 and slave SP 140 select that one slave SP 140 as an alternative path regardless of the reservation information.

  When the alternative path is formed, the synchronization unit 122 of the master SP 120 outputs the SP data (master SP data 161 and slave SP data 163a to 163n) to be transmitted to the standby SP 130 to the data control unit 123. The data control unit 123 transmits the SP data received from the synchronization unit 122 to the data passing unit 142 b of the selected slave SP 140 b. The data passing unit 142 b transmits the received SP data to the data control unit 133 of the standby SP 130. The data control unit 133 outputs the received SP data to the synchronization unit 132. The synchronization unit 132 stores the SP data received from the data control unit 133 in the storage unit 131.

  Similarly, the synchronization unit 132 of the standby SP 130 outputs the SP data (standby SP data 162) to be transmitted to the master SP 120 to the data control unit 133 of the standby SP 130. The data control unit 133 of the standby SP 130 transmits the SP data received from the synchronization unit 132 to the data passing unit 142 a of the selected slave SP 140 a. The data passing unit 142a transmits the received SP data to the data control unit 123 of the master SP 120. The data control unit 123 of the master SP 120 outputs the received SP data to the synchronization unit 122. The synchronization unit 122 stores the SP data received from the data control unit 123 in the storage unit 121.

  As described above, the master SP 120 or the standby SP 130, which is the SP data transmission source, detects a failure in the communication path used to transmit the SP data, and forms an alternative path. The master SP 120 and the standby SP 130 can store the same SP data as the transmission source SP forms an alternative path. By the master SP 120 and the standby SP 130 storing the same SP data, the standby SP 130 can maintain an operable state as a backup of the master SP 120. In addition, the slave SP 140, which is indicated by the reservation information that has already been selected as the alternative path, can be prevented from being double selected as the alternative path by being excluded from selection targets. . By preventing one slave SP 140 from being selected twice as a relay point of two alternative paths, it is possible to prevent an excessive load on the slave SP 140 to relay data transmission.

  FIG. 7 is a diagram for explaining an example of processing for forming an alternative path. In the BB system 100 of FIG. 7, the same components as in FIG. 2 are assigned the same reference numerals as in FIG. 2.

  In the BB system 100 of FIG. 7, a failure occurs in the communication path 151. Therefore, the standby SP 130 selects one slave SP 140 and starts forming an alternative path via the selected slave SP 140. Below, the formation process of an alternative path is shown in order.

  In step S1, the data control unit 133 of the standby SP 130 outputs, to the synchronization unit 132 in the own standby SP 130, a stop instruction for stopping the process of transmitting the SP data. By the stop instruction, data transmission from the synchronization unit 132 of the standby SP 130 to the synchronization unit 122 of the master SP 120 is stopped.

  In step S2, the data control unit 133 of the standby SP 130 stores the connection destination information set in the synchronization unit 132 in the storage unit 131 as the backup 166. The connection destination information set in the synchronization unit 132 includes an Internet Protocol address (IP) address or a port number of the information processing apparatus 101 corresponding to the communication path 151.

  In step S3, the data control unit 133 of the standby SP 130 sets the IP address and port number assigned to the data control unit 133 in the connection destination information of the synchronization unit 132.

  In step S4, the data control unit 133 of the standby SP 130 restarts the synchronization unit 132.

  By restarting, in step S5, the synchronization unit 132 of the standby SP 130 recognizes the IP address and port number set as connection destination information in step S3 and forms a communication path with the data control unit 133. The communication path between the synchronization unit 132 and the data control unit 133 is specifically a logical communication path for inter-process communication using a socket. Note that although the communication path 152 is disconnected once along with the restart of the synchronization unit 132, the communication path 152 is automatically rebuilt by a retry process or the like. That is, although the connection on the communication path 152 is disconnected once, since there is no failure in the physical cable for realizing the communication path 152, the connection on the communication path 152 is reestablished by a retry process or the like. The connection on the communication path 152 is specifically, for example, a Transmission Control Protocol (TCP) connection.

  In step S6, the data control unit 133 of the standby SP 130 refers to the via information 165 stored in the storage unit 131, and specifies the selected (reserved) slave SP 140.

  In step S7, the data control unit 133 of the standby SP 130 transmits an alternative path formation request to the data passing unit 142 of the selected slave SP 140.

  In step S8, the data passing unit 142 of the slave SP 140 sends, to the data control unit 123 of the master SP 120, connection destination information including the IP address and port number set in the data passing unit 142, and a request for forming an alternative path. Send. Here, the IP address of the data passing unit 142 is 192.168.200, and the port number is 1001. Therefore, in step S8, the data passing unit 142 of the slave SP 140 transmits connection destination information having an IP address of 192.168.200 and a port number of 1001 to the data control unit 123 of the master SP 120.

  In step S9, when the data control unit 123 of the master SP 120 receives the request to form an alternative path, it uses the connection destination information with an IP address of 192.168.200 and a port number of 1001 to communicate with the slave SP 140. Form a communication path. Specifically, the data control unit 123 establishes a connection between the data control unit 123 and the data relay unit 142.

  In step S10, the data control unit 123 of the master SP 120 forms a communication path with the synchronization unit 122. The communication path between the synchronization unit 122 and the data control unit 123 is specifically a logical communication path for inter-process communication using a socket.

  In step S11, the data control unit 123 of the master SP 120 transmits, to the data passing unit 142 of the slave SP 140, information indicating that the process corresponding to the request for forming the alternative path has been completed.

  In step S12, the data control unit 123 of the master SP 120 adds information indicating that the selected (reserved) slave SP 140 has already been selected as an alternative route to the via information 165 of the storage unit 121. The process of step S12 may be performed in parallel with step S11. Also, the process of step S12 may be performed before step S11.

  In step S13, the data passing unit 142 of the slave SP 140 transmits, to the data control unit 133 of the standby SP 130, information indicating that the process corresponding to the request for forming the alternative path has been completed on the data control unit 123 side of the master SP 120. . At the same time, the data passing unit 142 of the slave SP 140 transmits connection destination information including the IP address and port number set in the data passing unit 142 to the data control unit 133 of the standby SP 130. The data passing unit 142 transmits the connection destination information transmitted to the data control unit 123 of the slave SP 140 to a port number different from the connection destination information in step S8. For example, the connection destination information transmitted in step S13 has an IP address of the data passing unit 142 of 192.168.200 and a port number of 1000.

  In step S14, the data control unit 133 of the standby SP 130 forms a communication path with the slave SP 140 using the IP address and port number included in the received connection information. Specifically, the data control unit 133 establishes a connection between the data control unit 133 and the data relay unit 142.

  As described above, the data control unit 123 forms a communication path between the data passing unit 142 and the data control unit 123 of the master SP 120, and communication between the data passing unit 142 and the data control unit 133 of the standby SP 130. The data control unit 133 forms a path. As a result, the slave SP 140 (more specifically, the data relay unit 142) can relay transmission and reception of data between the master SP 120 and the standby SP 130.

  FIG. 7 is an example in the case where a failure occurs in the communication path 151. When a failure occurs in the communication path 152, the master SP 120 performs the same processing as steps S1 to S7 and S14, and the standby SP 130 performs the same processing as steps S8, S10, and S12. When a failure occurs in the communication path 152, the transmission in step S8 is replaced with the transmission from the slave SP 140 to the standby SP 130, and the transmission in step S13 is replaced with the transmission from the slave SP 140 to the master SP 120.

  FIG. 8 is a diagram for explaining an example of the route information. Master SP 120 via information 165 and slave SP 130 via information 166 are information on slave SP 140 with which the own SP can communicate, inquiry information indicating whether inquiry to slave SP 140 has been completed, via situation information, risk index value, Contains selection information. The example of the route information of FIG. 8 is an example of the route information 166 of the standby SP 130 shown in FIG.

  The via information in FIG. 8 includes information (for example, an identifier of each of the slave SPs 140a to 140n) indicating the slave SP 140a to 140n as the slave SP 140 with which the standby SP 130 can communicate. The route information 166 of the standby SP 130 includes, as information of the slave SP 140 with which the standby SP 130 can communicate, information indicating each of the slave SPs 140 a to 140 n mounted on the information processing apparatuses 103 a to 103 n detected by the detector 134. The route information 165 of the master SP 120 includes, as the information on the slave SP 140 with which the master SP 120 can communicate, information indicating each of the slave SPs 140 a to 140 n mounted on the information processing apparatuses 103 a to 103 n detected by the detection unit 124. Be

  The example of the route information in FIG. 8 includes inquiry information indicating whether or not each standby SP 130 has made an inquiry to each of the slave SPs 140 with which it can communicate. The inquiry information is information indicating whether or not the data control unit 133 of the standby SP 130 has already transmitted to the slave SP 140 a calculation instruction to index the slave SP data 163 using the coefficient information 164. When the data control unit 133 of the standby SP 130 has already sent the calculation instruction to the slave SP 140, “Done” indicating that the calculation instruction has been sent is set in the inquiry information. If a calculation instruction has not been sent to slave SP 140, the inquiry information is set to “not yet” indicating that the calculation instruction has not been sent. Also in the transit information 165 of the master SP 120, similarly, information indicating whether a calculation instruction is sent to the slave SP 140 is set in the inquiry information.

  The route information in FIG. 8 includes the route status information corresponding to each slave SP 140 with which the standby SP 130 can communicate. The passing situation information is information indicating the communication situation of each slave SP 140. The passage status information is indicated by values such as “via OK”, “via NG”, “via via reservation”, “via via”, “unknown” and the like. The passing situation information is received from the slave SP 140 in accordance with the calculation instruction of the risk index value. "By way OK" indicates that the slave SP 140 can communicate with both the master SP 120 and the standby SP 130. "By way NG" indicates that the slave SP 140 can not communicate with at least one of the master SP 120 and the standby SP 130. The “via-reserved” in the via information 166 of the standby SP 130 indicates that the slave SP 140 has received a request for forming an alternative route from the master SP 120 side. The “via-reserved” in the via information 165 of the master SP 120 indicates a state in which the slave SP 140 has received a request for forming an alternative path from the standby SP 130 side. "By way through" indicates that the slave SP 140 is already used as an alternative path. "Unknown" indicates that whether or not the slave SP 140 can communicate with the master SP 120 and the standby SP 130 is unknown because it is unconfirmed. “Unknown” also indicates that the state of a hardware device such as a processing device mounted on the slave SP 140 is unknown.

  The risk index value is a value obtained by indexing the factor that adversely affects data transfer from the slave SP data 163 using the information included in the coefficient information 164 of the storage unit 141 by the data passing unit 142 of the slave SP 140 (see FIG. The details will be described later with reference to FIG. The risk index value is transmitted from the slave SP 140 to the standby SP 130 and written to the storage unit 131 as part of the route information 165. The risk index value is a value obtained by indexing factors that adversely affect data transfer, so the smaller the value (the less adverse effect), the better. In the route information 166 of FIG. 8, the risk index value for slave SP 140a is 1, the risk index value for slave SP 140b is 25, and the risk index value for slave SP 140n is 20. Then, the slave SP 140a with the smallest risk index value has few factors that adversely affect data transfer, and it can be determined that it is desirable to relay data. Therefore, the data control unit 123 of the master SP 120 and the data control unit 133 of the standby SP 130 select the slave SP 140 having the smallest value of the risk index value when selecting the slave SP 140 used to form the alternative path.

  The selection information is information indicating whether each slave SP 140 is selected as an alternative path. The data control unit 123 of the master SP 120 and the data control unit 133 of the standby SP 130 set “selection” in the selection information corresponding to the slave SP 140 used for forming the alternative path. In addition, "not selected" is set to the selection information corresponding to slave SP 140 which is not used as an alternative path.

  FIG. 9 is a diagram for explaining an example of coefficient information. The coefficient information 164 is preset in the storage unit 141 of the slave SP 140. The coefficient information 164 is used to index the state of the self information processing apparatus 103 in which the slave SP 140 is mounted.

  The slave SP 140 manages hardware devices such as a processor (processing device) and a memory in the own information processing device. In practice, the slave SP 140 includes a management unit 143 that manages hardware devices such as processing devices. When receiving the calculation instruction from the data control unit 123 of the master SP 120 or the data control unit 133 of the standby SP 130, the data passing unit 142 inquires of the management unit 143 the state of each hardware device. When data passing unit 142 receives information indicating the state of each processing device from management unit 143 (that is, at least a part of slave SP data 163), data passage unit 142 receives the information received from management unit 143 and coefficient information 164 in storage unit 141. To calculate the risk index value.

Here, the coefficient information 164 is information indicating the state of a hardware device such as a processing device, and an index value that indexes the degree to which the state indicated by the state information adversely affects data transmission via the slave SP 140. including. In other words, each index value is a coefficient corresponding to each state. There are the following examples of the state of the hardware device.
-Partition ON is in progress (during high load processing)
-Partition ON is in progress (low load processing)
-Partition ON state-Partition OFF in progress-Partition OFF state-Some hardware devices have failed-I / O devices are connected-Four information processing devices belong to one partition-Three information processing Device belongs to one partition-Two information processing devices belong to one partition-One information processing device belongs to one partition

  Index values are set in the coefficient information 164 corresponding to each of the above-described state information of the hardware device. One or more information processing apparatuses belong to one partition, and one operating system (OS) is executed on one partition. Specifically, one OS is executed by one or more processing devices mounted on one or more information processing apparatuses belonging to one partition. Thus, one information processing apparatus is logically realized by one or more information processing apparatuses belonging to one partition.

  Partition ON is processing for causing one or more information processing apparatuses to logically operate as one information processing apparatus, and processing for forming partitions logically. The partition ON execution (preparation) process is divided into a high load period and a low load period. The high load period adversely affects data transmission than the low load period. Therefore, in the coefficient information 164 of FIG. 9, the index value of “partition ON in progress (high load processing)” is 50, and the index value of “partition ON in progress (low load processing)” is 40. In the "partition ON state", since the environment for operating one or more information processing apparatuses logically as one information processing apparatus is already established, the process of partition setting for the information processing apparatus is not performed. There are fewer factors that adversely affect data transmission in "partition ON state" than "partition ON in progress". Therefore, the index value 30 is set in the "partition ON state".

  “Partition OFF in progress” indicates that processing for stopping the environment which has logically been operated as one information processing apparatus is being executed. In “partition off in progress”, the OS is stopped, so there are fewer factors that adversely affect data transmission than in the “partition on state” in which the OS is running. Therefore, the index value 20 is set in "partition OFF in progress". In the “partition off state”, since the partition off process has already been completed, each information processing apparatus is not in a process related to the partition. The "partition off state" has less factors that adversely affect data transmission than "partition off in progress". Therefore, an index value of 10 is set in the “partition OFF state”.

  If a part of the hardware devices of the processing devices included in the information processing device 103 in which the slave SP 140 is mounted is broken, it is conceivable that data transmission via the slave SP 140 is adversely affected. Therefore, 100 is set as the index value in the case where some hardware devices fail. When an I / O device is connected to the information processing device 103 in which the slave SP 140 is mounted, there is a possibility that the I / O device in the slave SP 140 may be used for another process, so data transmission of slave SP 140 is performed. It is possible that the processing is adversely affected. Therefore, 20 is set as the index value when the I / O device is connected.

  In the coefficient information 164, index values are further set according to the number of information processing apparatuses belonging to the partition. For example, when the number of information processing apparatuses belonging to a partition to which an information processing apparatus on which a certain processing apparatus is mounted belongs is 3, state information of the processing apparatus indicates that "three information processing apparatuses belong to one partition Is represented. For example, when there are four information processing devices, the probability that a failure occurs in at least one of the four information processing devices is about four times that in the case where one information processing device is present. Therefore, when there are four information processing apparatuses belonging to one partition, the index value is set to 40. If one information processing apparatus belongs to one partition, the index value is set to 10.

  In addition, the value set to said index value is changeable, and does not limit a numerical value. Further, the state of a hardware device such as a processing device may further include information indicating a further different state.

  The data passing unit 142 calculates a risk index value by adding together index values corresponding to the state of the processing device represented by the slave SP data 163. When the data control unit 123 of the master SP 120 and the data control unit 133 of the slave SP 130 receive the risk index value from the data passing unit 142, they substitute the information processing apparatus 103 (more specifically, the slave SP 140) having a small risk index value. Choose for route formation.

  FIG. 10 is a flow chart for explaining an example of the alternative route formation processing of the master SP. If there is a failure in the communication path 152 used by the synchronization unit 122 to transmit SP data to the synchronization unit 132, the detection unit 124 detects that there is a failure (step S101). If there is no failure in the communication path 152 (NO in step S101), the detection unit 124 repeats the process of step S101 (that is, monitoring of the communication path 152). If there is a failure in the communication path 152 (YES in step S101), the detection unit 124 notifies the data control unit 123 that a failure has occurred (step S102).

  The data control unit 123 refers to the route information 165 and determines whether there is a slave SP 140 that can communicate with the master SP 120 (step S103). If there is no slave SP 140 that can communicate with the master SP 120 (NO in step S 103), the master SP 120 ends the alternative path formation process. If there is one or more slave SP 140 that can communicate with master SP 120 (YES in step S 103), data control unit 123 transmits a calculation instruction to calculate a risk index value to all communicable slave SPs 140 (step S 104). ).

  The data control unit 123 determines whether the risk index value has been received from all the communicable slaves SP 140 (step S105). If the risk index value has not been received from all communicable slaves SP 140 (NO in step S 105), the data control unit 123 repeats the process of step S 105. When the risk index value is received from all communicable slaves SP 140 (YES in step S 105), the data control unit 123 selects the slave SP 140 having the smallest risk index value (step S 106).

  The data control unit 123 forms an alternative path using the selected slave SP 140 (step S107). After forming the alternative path, the data control unit 123 ends the alternative path formation process. When the alternative path is formed, the synchronization unit 122 of the master SP 120 transmits the SP data to the data control unit 133 of the standby SP 130 via the data control unit 123 and the alternative path.

  FIG. 11 is a flow chart for explaining an example of a standby SP alternative path formation process. If there is a failure in the communication path 151 used by the synchronization unit 132 to transmit SP data to the synchronization unit 122, the detection unit 134 detects that there is a failure (step S201). If there is no failure in the communication path 151 (NO in step S201), the detection unit 134 repeats the process of step S201 (that is, monitoring of the communication path 151). If there is a failure in the communication path 151 (YES in step S201), the detection unit 134 notifies the data control unit 133 that a failure has occurred (step S202).

  The data control unit 133 refers to the route information 166 and determines whether there is a slave SP 140 that can communicate with the standby SP 130 (step S203). If there is no slave SP 140 that can communicate with the standby SP 130 (NO in step S 203), the standby SP 130 ends the alternative path formation processing. If there is one or more slave SPs 140 that can communicate with the standby SP 130 (YES in step S203), the data control unit 133 transmits a calculation instruction for calculating the risk index value to all the communicable slaves SP 140 (step S204). ).

  The data control unit 133 determines whether the risk index value has been received from all communicable slaves SP 140 (step S205). If the risk index value has not been received from all communicable slaves SP 140 (NO in step S 205), the data control unit 133 repeats the process of step S 205. When the risk index value is received from all communicable slaves SP 140 (YES in step S 205), the data control unit 133 selects the slave SP 140 having the smallest risk index value (step S 206).

  The data control unit 133 forms an alternative path using the selected slave SP 140 (step S207). The details of step S207 are as described with reference to FIG. After forming the alternative path, the data control unit 133 ends the alternative path formation process. When an alternative path is formed, the synchronization unit 132 of the standby SP 130 transmits SP data to the data control unit 123 of the master SP 120 via the data control unit 133 and the alternative path.

  12A and 12B are flowcharts illustrating an example of the process of calculating the risk index value of slave SP. Upon receiving the calculation instruction for calculating the risk index value from the data control unit 123 of the master SP 120 or the data control unit 133 of the standby SP 130, the data passing unit 142 starts the calculation process of FIGS. 12A to 12B.

  The detection unit 144 determines whether the slave SP 140 can communicate with both the master SP 120 and the standby SP 130 (step S301). When slave SP 140 can communicate with both master SP 120 and standby SP 130 (YES in step S 301), data relay unit 142 inquires of management unit 143 the status of each hardware device installed in information processing apparatus 103 (step S302). The data passing unit 142 receives slave SP data 163 indicating the state of each hardware device from the managing unit 143. Then, the data passing unit 142 determines whether the index information corresponding to the state of the hardware device represented by the received slave SP data 163 is included in the coefficient information 164 (step S303). The data passing unit 142 determines, for at least one hardware device, whether the coefficient information 164 includes an index value corresponding to the state of the hardware device. When the index value corresponding to the state of the processing device is not included in the coefficient information 164 (NO in step S303), the data passing unit 142 sends the calculation instruction to the data control unit 123 or the data control unit 133 which is the transmission source. Information indicating "unknown" is transmitted (step S304). If slave SP 140 can not communicate with at least one of master SP 120 and standby SP 130 (YES in step S 301), data relay unit 142 sends a “NG via” to data control unit 123 or data control unit 133 that is the transmission source of the calculation instruction. Is transmitted (step S305). When the process of step S304 or step S305 is completed, the slave SP 140 ends the process of calculating the risk index value.

  When the index value corresponding to the state of at least one or more hardware devices is included in the state information 164 (YES in step S303), the data passing unit 142 adds all index values corresponding to the state of the hardware device. Thus, the risk index value is calculated (step S306). When there are a plurality of hardware devices, the risk index value corresponding to the state of each hardware device is further summed. The data passing unit 142 determines whether the own information processing apparatus 103 equipped with the slave SP 140 including the data passing unit 142 is already under reservation or passing (step S307). Note that information such as “via via reservation” and “via via” is held as internal data of the data via unit 142.

  When the information processing apparatus 103 is in transit reservation or not in transit (NO in step S307), the data transit unit 142 transmits the risk index value and the reservation information indicating “pass through OK” to the transmission source of the calculation instruction. It transmits to the control part 123 or the data control part 133 (step S308). When the information processing apparatus 103 is in transit reservation or in transit (YES in step S307), the data transit unit 142 determines whether the information processing apparatus 103 equipped with the slave SP 140 including the data transit unit 142 is in transit. Is determined (step S309). When the information processing apparatus 103 is not passing through (NO in step S309), the data passing unit 142 transmits the calculation instruction, which is the transmission source of the calculation instruction, to the risk index value and the reservation information indicating "via reservation". It transmits to the data control unit 133 (step S310). When the information processing apparatus 103 is in transit (YES in step S309), the data transit unit 142 transmits the risk index value and the reservation information indicating “in transit” to the data control unit 123 or the data that is the transmission source of the calculation instruction. It transmits to the control part 133 (step S311). When the process of step S308, step S310 or step S311 is completed, the slave SP 140 ends the process of calculating the risk index value.

  13A and 13B are flowcharts illustrating an example of slave SP selection processing. As shown in FIGS. 12A and 12B, the risk index value and the reservation information indicating the via state are transmitted from the slave SP 140 to the data control unit 123 of the master SP 120 or the data control unit 133 of the standby SP 130. Upon receiving the risk index value and the reservation information, the data control unit 123 of the master SP 120 or the data control unit 133 of the standby SP 130, according to the received risk index value and reservation information, the passing situation information and the risk index value in the via information 165. Update After updating the passing information 165, the data control unit 123 or the data control unit 133 starts the selection process of FIGS. 13A and 13B. Hereinafter, for convenience of description, a case where the data control unit 123 of the master SP 120 starts the selection process of the slave SP 140 will be specifically described.

  The data control unit 123 of the master SP 120 refers to the via information 165, and determines whether there is a slave SP 140 of via OK (step S401). When the entry for the via OK slave SP 140 is included in the via information 165 (YES in step S 401), the data control unit 123 refers to the via information 165 and the numerical value of the risk index value in the via OK slave SP 140 is The smallest slave SP 140 is selected (step S402). The data control unit 123 transmits an alternative path formation request to the data passing unit 142 of the selected slave SP 140 (step S403).

  For example, as illustrated in FIG. 6, when a failure occurs in the communication path 151 and the communication path 152 almost simultaneously, the calculation instruction of the risk index value is transmitted from the data control unit 123 and the data control unit 133 almost simultaneously. It is sent to 142. In this case, the risk index value and the reservation information “via via” can be transmitted from the data transit unit 142 to each of the data control unit 123 and the data control unit 133 almost simultaneously. Further, in this case, the request for forming the alternative route described in regard to step S7 of FIG. That is, each of the data control unit 123 and the data control unit 133 may execute the process of step S403 almost simultaneously. In this case, the data relay unit 142 performs the processes in steps S8 to S13 of FIG. 7 in accordance with the formation request received earlier.

  For example, when the data relay unit 142 receives the formation request from the data control unit 133 first, the data relay unit 142 performs data control that the process corresponding to the formation request has been completed, as shown in step S13 of FIG. 7. It notifies part 133. This notification is, in other words, a response "OK" to the formation request. When the data relay unit 142 receives the formation request from the data control unit 133 first, the data relay unit 142 returns a response of “NG” to the data control unit 123. The response “NG” indicates that the data relay unit 142 can not answer the formation request from the data control unit 123 because the data relay unit 142 is already “via reservation”.

  Steps S403 to S406 are steps for forming two different alternative paths when failures occur in the communication path 151 and the communication path 152 almost simultaneously. Specifically, the data control unit 123 determines whether the response from the data passing unit 142 of the slave SP 140 selected in step S402 is OK (step S404). If it is possible to form an alternative path passing through the selected slave SP 140 (YES in step S 404), the data control unit 123 selects “select” in the selection information corresponding to the selected slave SP 140 in the transit information 165. It sets (step S405). If the alternative path through the selected slave SP 140 can not be formed (NO in step S 404), the data control unit 123 sets the transit status information corresponding to the selected slave SP 140 in the transit information 165 to “via via reservation”. The change is made (step S406). When step S406 ends, the data control unit 123 repeats the process from step S401.

  If the entry for the via OK slave SP 140 is not included in the via information 165 (NO in step S 401), the data control unit 123 refers to the via information 165, and the slave SP 140 is set to be via reservation or via. It is determined whether there is any (step S407). If there is no slave SP 140 that is set as being reserved or via in the via information 165 (NO in step S 407), the data control unit 123 determines that an alternative route via the slave SP 140 can not be formed. The selection process is ended (step S408).

  If there is a slave SP 140 that is set as being in transit reservation or in transit in the transit information 165 (YES in step S 407), the data control unit 123 selects the slave SP 140 which is set in transit reservation or in transit ( Step S409). It should be noted that there is only one slave SP 140 which is set to be in transit reservation or in transit reservation. When the process of step S409 ends, the data control unit 123 starts the process from step S405. When there is a failure in the communication path 151, the data control unit 133 of the standby SP 130 executes the same process as the process of steps S401 to S410 in FIGS. 13A and 13B.

  FIG. 14 is a flowchart illustrating an example (part 1) of the process of forming an alternative path. The data control unit 123 of the master SP 120 transmits, to the synchronization unit 122, an instruction to stop the process of transmitting the SP data (step S501). The synchronization unit 122 stops the transmission process of the SP data (step S502). The data control unit 123 backs up the connection destination information including the IP address and the port number set in the synchronization unit 122 to the storage unit 121 (step S503). The data control unit 123 sets the IP address and port number assigned to the data control unit 123 in the connection destination information of the synchronization unit 122 (step S504).

  The data control unit 123 transmits an instruction to restart the synchronization unit 122 to the synchronization unit 122 (step S505). Then, the synchronization unit 122 restarts (step S506). The synchronization unit 122 recognizes the IP address and port number set as connection destination information (that is, the IP address and port number assigned to the data control unit 123) by restarting, and the synchronization unit 122 performs communication with the data control unit 123. Form a communication path (step S507). Steps S501 to S507 are similar to steps S1 to S5 of FIG.

  The data control unit 123 refers to the route information 165 and specifies the selected (reserved) slave SP 140 (step S 508). The data control unit 123 transmits an alternative path formation request to the selected slave SP 140 (step S509). Steps S508-S509 are similar to steps S6-S7 of FIG.

  The data control unit 123 determines whether the information indicating that the process corresponding to the alternative path formation request has been completed has been received from the data transit unit 142 of the slave SP 140 (step S510). If the information indicating that the process corresponding to the request for formation of the alternative path has been completed is not received from data passing unit 142 of slave SP 140 (NO in step S510), data control unit 123 repeats the process of step S510. .

  As described in regard to steps S403 to S406 in FIG. 13A, in rare cases, the data control unit 123 may receive a response of “NG” from the data passing unit 142. However, for simplification of the drawing, this case is omitted in FIG. If the response “NG” is received, the data control unit 123 ends the process of FIG.

  On the other hand, when the information indicating that the process corresponding to the request for formation of the alternative path is completed is received from data passing unit 142 of slave SP 140 (YES in step S510), data control unit 123 A communication path is formed (step S511). The transmission of the information indicating that the process corresponding to the request for formation of the alternative path has been completed is similar to step S13 in FIG. Further, this information is a response “OK” to the formation request (see step S404 in FIG. 13A). Step S511 is similar to step S14 of FIG.

  The data control unit 123 transmits an instruction to start synchronization processing of SP data to the synchronization unit 122 (step S512). The synchronization unit 122 of the master SP 120 starts transmission of SP data to the synchronization unit 132 of the standby SP 130 via the data passing unit 142 (step S513). The data control unit 123 ends the process of forming an alternative path. When a failure occurs in the communication path 151, the data control unit 133 and the synchronization unit 132 of the standby SP 130 execute the same processes as the processes of S501 to S513 in FIG.

  FIG. 15 is a flowchart illustrating an example (part 2) of the process of forming an alternative path. The process of FIG. 15 is an example of the process executed by the slave SP 140 which has received the request for forming the alternative path in step S 509 of FIG. 14 from the master SP 120 side. Also when slave SP 140 receives a request for forming an alternative path in step S 509 of FIG. 14 from standby 130, slave SP 140 executes processing similar to that of FIG. 15 (see also FIG. 7).

  The data passing unit 142 transmits, to the data control unit 133 of the standby SP 130, connection destination information including the IP address and Port number set in the data passing unit 142 and a request for forming an alternative path (step S601). Step S601 is similar to step S8 of FIG. The data passing unit 142 determines whether the alternative path formation process on the standby SP 130 side is completed (step S602). For example, the data passing unit 142 receives, from the data control unit 133, information (information similar to the information transmitted in step S11 of FIG. 7) indicating that the process corresponding to the request for forming the alternative path has been completed. Determine if

  If the process of forming an alternative path on the standby SP 130 side has not been completed (NO in step S602), the data passing unit 142 repeats the process of step S602. When the formation of the alternative path on the standby SP 130 side is completed (YES in step S602), the data relay unit 142 notifies the data control unit 123 of the master SP 120 that the process of forming the alternative path on the slave SP 130 side is completed. It transmits (step S603). The notification in step S603 is similar to the notification in step S13 of FIG. In step S603, the data passing unit 142 also transmits connection destination information including the IP address and port number set in the data passing unit 142 to the data control unit 123 of the master SP. When the process of step S603 is completed, the process of step S510 of FIG. 14 is started.

  FIG. 16 is a flowchart illustrating an example (part 3) of the process of forming an alternative path. The process of FIG. 16 is an example of the process on the side of the standby SP 130 after receiving the request for forming an alternative path in step S601 of FIG.

  When the data control unit 133 of the standby SP 130 receives the formation request of the alternative path together with the connection destination information, the data control unit 133 forms a communication path with the slave SP 140 using the IP address and port number included in the received connection destination information (step S701). The data control unit 133 forms a communication path with the synchronization unit 132 (step S702). The data control unit 133 transmits, to the data passing unit 142 of the standby SP 130, information indicating that the process corresponding to the request for forming the alternative path (that is, the request transmitted in step S601) is completed (step S703). Steps S701 to S703 are similar to steps S9 to S11 of FIG. When the process of step S703 is completed, the process proceeds to step S602 of FIG. When there is a failure in the communication path 151, the master SP 120 executes a process similar to steps S701 to S703 in FIG. 16 (see steps S9 to S11 in FIG. 7).

  FIG. 17 is a flowchart for explaining an example of processing after the failure of the communication path is recovered. The detection unit 124 detects the recovery of the failure of the communication path 152 (step S801). The detection unit 124 transmits, to the data control unit 123, information indicating that the failure of the communication path 152 has been recovered (step S802). The data control unit 123 refers to the route information 165 and determines whether the slave SP 140 is used as an alternative route (step S 803). If the slave SP 140 is not used as an alternative path (NO in step S 803), the data control unit 123 ends the process of FIG. When slave SP 140 is used as an alternative path (YES in step S 803), data control unit 123 transmits a stop instruction to stop transmission of SP data to synchronization unit 122 (step S 804).

  The synchronization unit 122 stops transmission of the SP data (step S805). The data control unit 123 resets the connection destination information set in the synchronization unit 122 in step S504 in FIG. 14 (step S806). The data control unit 123 refers to the connection destination information backed up in the storage unit 121 in step S 503 of FIG. 14 and sets the IP address and port number of the synchronization unit 132 of the standby SP 130 as the connection destination of the synchronization unit 122 (step S807). The data control unit 123 transmits an instruction to restart the synchronization unit 122 to the synchronization unit 122 (step S808).

  The synchronization unit 122 restarts (step S809). The synchronization unit 122 forms a communication path with the synchronization unit 132 of the standby SP 130 (step S810). The data control unit 123 ends the processing after the failure of the communication path is recovered. The recovery of the failure of the communication path 151 is detected by the detection unit 134. The standby SP 130 performs processing similar to that of FIG.

  FIG. 18 is a flowchart illustrating an example of the process of recalculating the risk index value. Each slave SP 140 preferably recalculates the risk index value periodically even after transmitting the calculation result of the risk index value to the master SP 120 or the standby SP 130. The risk index value of slave SP 140 is not constant. Therefore, when the risk index value of the selected slave SP 140 becomes higher than a predetermined threshold, it is desirable to use another slave SP 140 as an alternative path. The processing of the slave SP 140 shown in FIG. 18 is preferably performed periodically.

  The data relay unit 164 acquires the status information of the processing apparatus (specifically, at least a part of the slave SP data 163) from the management unit 143, and calculates a risk index value (step S901). The data relay unit 164 determines whether the calculated risk index value is higher than a predetermined threshold (step S902). If the calculated risk index value is lower than the predetermined threshold (NO in step S902), the data relay unit 164 ends the recalculation process of the risk index value. If the calculated risk index value is higher than the predetermined threshold (YES in step S 902), data passing unit 164 transmits the reselection command of slave SP 140 to the data control unit of the transmission source of the risk index value calculation command. (Step S903). The data relay unit 164 ends the recalculation process.

  When the data control unit 123 receives the reselection instruction, the data control unit 123 executes the processes of steps S104 to S107 of FIG. When the data control unit 133 receives the reselection instruction, the data control unit 133 executes the processes of steps S204 to S207 of FIG.

  As described above, even if a failure occurs in the communication path 152 for transmitting SP data from the master SP 120 to the standby SP 130, the master SP 120 can transmit SP data to the standby SP 130 via the slave SP 140. Further, even if a failure occurs in the communication path 151 for transmitting SP data from the standby SP 130 to the master SP 120, the standby SP 130 can transmit SP data to the master SP 120 via the slave SP 140. By the master SP 120 and the standby SP 130 storing the same SP data, the standby SP 130 can maintain an operable state as a backup of the master SP 120.

100 BB systems 101, 102, 103a to 103n, 200 information processing apparatus 110 I / O apparatus 120 master SP
121, 131, 141a to 141n storage unit 122, 132, 145a to 145n synchronization unit 123, 133 data control unit 124, 134, 144a to 144n detection unit 130 standby SP
140a to 140n slave SP
142a to 142n Data transit units 143a to 143n Management units 151, 152, 153a to 153n, 154a to 154n Communication path 161 Master SP data 162 Standby SP data 163a to 163n Slave SP data 164 Coefficient information 165, 166 Via information 210 CMU
220 SP
211, 221, 11 processor 212, 222, 12 memory

Claims (8)

An information processing system including a plurality of processing devices and a plurality of management devices managing the plurality of processing devices, the information processing system comprising:
A first management device for storing data;
A second management device that receives the data from the first management device via a communication path and stores the received data;
And one or more other management devices,
When a failure occurs in the communication path, the first management device performs the first management among the one or more other management devices based on the state of the one or more other management devices. A device and a third management device connected to the second management device, and transmitting the data to the second management device via the third management device ;
The second management device transmits second data stored by the second management device to the first management device via the communication path or another communication path.
If a failure occurs in the communication path used by the second management device for transmitting the second data or the other communication path, the second management device determines the state of the one or more other management devices. And select a fourth management device different from the third management device or the third management device among the one or more other management devices, and select the third management device Or transmitting the second data to the first management device via the selected fourth management device;
The state of each of the other one or more management devices is an indicator based on the degree to which the state of one or more processing devices managed by the other management device adversely affects data transmission via the other management device. Represented by a value,
The first management device selects the third management device based on the index value,
An information processing system, wherein the second management apparatus selects the third management apparatus or the fourth management apparatus based on the index value . The first management device causes the third management device to transmit, to the first management device, first connection destination information including an Internet Protocol (IP) address and a port number of the third management device. Causing the second management apparatus to transmit second connection destination information including a port number different from the IP address and the port number;
The second management device forms a communication path between the second management device and the third management device according to the second connection destination information.
The first management apparatus, according to the first connection destination information, according to claim 1, characterized in that to form a communication path between the first management apparatus and the third management apparatus Information processing system. When the second management device selects the third management device, the second management device receives first connection destination information including an Internet Protocol (IP) address and a port number of the third management device, the third management device. Transmission to the first management device and the second management device, and forming a communication path between the second management device and the third management device according to the first connection destination information ,
When the second management apparatus selects the third management apparatus, the first management apparatus performs the first management apparatus and the third management apparatus according to the first connection destination information. Form a communication path between
When the second management device selects the fourth management device, the second connection destination information including an IP address and a port number of the fourth management device is transmitted from the fourth management device to the first connection destination information. Forming a communication path between the second management apparatus and the fourth management apparatus according to the second connection destination information.
When the second management apparatus selects the fourth management apparatus, the first management apparatus performs the first management apparatus and the fourth management apparatus according to the second connection destination information. The information processing system according to claim 1 , forming a communication path between the two. In a first management device of an information processing system including a plurality of processing devices and a plurality of management devices managing the plurality of processing devices,
Detecting a fault occurring in a communication path between a second management apparatus that receives data stored by the first management apparatus from the first management apparatus, and the first management apparatus;
When a failure occurs in the communication path, the first management device and the second management device are selected from among the one or more other management devices based on the state of the other one or more management devices. Select the third management device connected,
Transmitting the data to the second management device via the third management device ;
The state of each of the other one or more management devices is an indicator based on the degree to which the state of one or more processing devices managed by the other management device adversely affects data transmission via the other management device. Represented by a value,
Allowing the first management device to further execute a process of causing the third management device to be selected based on the index value
A program characterized by The second management apparatus transmits first connection destination information including the Internet Protocol (IP) address and port number of the third management apparatus from the third management apparatus to the second management apparatus. After the communication path between the second management device and the third management device is formed according to the first connection destination information, the IP address and the IP address are transmitted from the third management device to the first management device. Transmitting an instruction to transmit second connection destination information including a port number different from the port number to the third management device;
Receiving the second connection destination information from the third management device;
The program according to claim 4 , wherein the first management apparatus further executes a process of forming a communication path with the third management apparatus according to the second connection destination information. An information processing method executed in an information processing system including a plurality of processing devices and a plurality of management devices managing the plurality of processing devices,
When a failure occurs in a communication path between a first management device storing data and a second management device receiving the data from the first management device, the first management device performs another operation. Select a third management device connected to the first management device and the second management device from among the one or more other management devices based on the state of one or more management devices Sending the data to the second management device via the third management device ;
The second management device transmits second data stored by the second management device to the first management device via the communication path or another communication path;
If a failure occurs in the communication path used by the second management device for transmitting the second data or the other communication path, the second management device determines the state of the one or more other management devices. And select a fourth management device different from the third management device or the third management device among the one or more other management devices, and select the third management device Or transmitting the second data to the first management device via the selected fourth management device;
The state of each of the other one or more management devices is an indicator based on the degree to which the state of one or more processing devices managed by the other management device adversely affects data transmission via the other management device. Represented by a value,
The first management device selects the third management device based on the index value,
An information processing method characterized in that the second management apparatus selects the third management apparatus or the fourth management apparatus based on the index value . The first management device causes the third management device to transmit, to the first management device, first connection destination information including an Internet Protocol (IP) address and a port number of the third management device. Transmitting second connection destination information including a port number different from the IP address and the port number to the second management device;
The second management device forms a communication path between the second management device and the third management device according to the second connection destination information.
The first management apparatus, wherein according to said first connection destination information, in claim 6, characterized in that to form a communication path between the first management apparatus and the third management apparatus Information processing method. When the second management apparatus selects the third management apparatus, the first connection destination information including the Internet Protocol (IP) address and port number of the third management apparatus is transmitted to the third management apparatus. Transmission to the first management device and the second management device, and forming a communication path between the second management device and the third management device according to the first connection destination information ,
When the second management apparatus selects the third management apparatus, the first management apparatus performs the first management apparatus and the third management apparatus according to the first connection destination information. Form a communication path between
When the second management device selects the fourth management device, second connection destination information including an IP address and a port number of the fourth management device is transmitted from the fourth management device to the first connection destination information. Forming a communication path between the second management apparatus and the fourth management apparatus according to the second connection destination information.
When the second management apparatus selects the fourth management apparatus, the first management apparatus performs the first management apparatus and the fourth management apparatus according to the second connection destination information. The information processing method according to claim 6 , forming a communication path between the two.

Images