WO2011138886A1

WO2011138886A1 - Fault tolerant computer system and data access method

Info

Publication number: WO2011138886A1
Application number: PCT/JP2011/058576
Authority: WO
Inventors: 山本　秀典; 加藤　博光; 足達　芳昭
Original assignee: 株式会社日立製作所
Priority date: 2010-05-06
Filing date: 2011-04-05
Publication date: 2011-11-10
Also published as: JP2011237846A; JP5331749B2

Abstract

Provided in a fault tolerant computer system is a method for a user to conduct command execution or data editing with respect to a node that constitutes the system, without being conscious of the system configuration or data arrangement. When the user is to refer to or edit the same data that a plurality of processing nodes constituting the fault tolerant computer system possess, the applicable data is collected from each of the processing nodes into a maintenance node, the content thereof are compared, and the content is displayed on the basis of the result of the comparison. When the user edits the data, the maintenance node transmits the edited data to the plurality of processing nodes, and overwrites and updates the data all at once.

Description

Fault tolerant computer system and data access method

The present invention relates to a fault tolerant computer (Fault Tolerant Computer) system, which is composed of a plurality of independent nodes interconnected via a network, and allows the same processing to be executed in parallel in each of the nodes constituting the system. The present invention relates to a tool used when a user performs work such as debugging and testing of an application program running on a fault tolerant computer system.

Patent Documents 1 to 3 disclose various techniques related to data access processing for a plurality of devices.

In Patent Document 1, in order to edit a file on a remote workstation, the file is transferred / stored to the local workstation and then edited locally, and the edited file is transferred to the original remote workstation and stored. A remote editor support apparatus and method is disclosed.

In Patent Document 2, an area for storing own computer data and an area for storing partner computer data are prepared in a multiplexed computer, and when copying, data and mode information (control / standby / A method of using data between multiplexed computers is disclosed, in which the mode of the own computer and the partner computer is compared and the data in the mode designated in advance is adopted.

Patent Document 3 includes access authority information between terminals in a distributed file system including a plurality of terminal devices that manage files by storing files that collectively handle data under a directory that indicates the position of a hierarchical structure. A distributed file sharing system and its file access control method are disclosed in which index information is shared and referred to during search and access to minimize processing time.

JP-A-9-231155 Japanese Patent Laid-Open No. 2003-162441 JP 2002-132530 A

In a fault-tolerant computer, which is composed of a plurality of independent nodes interconnected via a network, and the same processing is executed in parallel in each of the constituent nodes, a plurality of parallel processing is performed to guarantee the operation of the system. Data used by applications for processing between nodes must always have the same contents. For this reason, when the user edits these data, it is necessary to be careful not to cause a difference in contents between nodes.

However, in the technique disclosed in Patent Document 1, for example, when referring to and editing the same data in a plurality of processing nodes constituting the fault tolerant computer system, it is necessary to operate each node and access the data. Work load increases. It can also cause human error. For example, in the technique disclosed in Patent Document 2, when referring to and editing the same data in a plurality of processing nodes constituting a fault-tolerant computer system, it is necessary to individually access and operate a directory mapped to each node. , User workload increases. It can also cause human error. For example, in the technique disclosed in Patent Document 3, data is identified between computers by duplication regardless of data contents, but when executed between processing nodes constituting a fault-tolerant computer system, it differs from data that should have the same contents. It is necessary to distinguish data (logs, node-specific information such as IP addresses, etc.) that is the contents.
The present invention takes the above-described problems into consideration, and in the operations such as debugging and testing, the user executes command and data on the nodes constituting the system without being aware of the system configuration and data arrangement. An object of the present invention is to provide a data access method in a fault tolerant computer system for editing and the like.

In order to solve the above problems, the present invention has the following configuration. That is, a method in a fault tolerant computer system including a plurality of processing nodes connected via a network and a maintenance node that acquires, edits, and updates data in the plurality of processing nodes, the plurality of processing nodes being the same When the user refers to or edits the same data that the plurality of processing nodes have, the maintenance node performs the same data in the plurality of processing nodes, respectively. The step of acquiring and saving from the processing node via the network and the maintenance node comparing the contents of the same data acquired from the plurality of processing nodes, and displaying the contents to the user via the user terminal based on the comparison result And pre-defined on the maintenance node via the user terminal A step in which a user edits data acquired from one processing node that matches a condition specified in advance in a processing node that is a table or an operating processing node, and the maintenance node is after the editing Transferring the data to the plurality of processing nodes, and overwriting and updating.

According to the present invention, when a user performs debugging, testing, or the like of an application program, the load of operation by the user regardless of the complexity of the configuration of the system, software, etc. Is reduced and work efficiency is improved.

It is a figure which shows the outline | summary of the data access method in a fault tolerant computer system. It is a figure which shows the outline | summary of the fault tolerant computer system comprised by the some independent node mutually connected via the network (communication medium), and performing the same process in parallel in each node which comprises. It is a figure which shows the whole process sequence between several nodes which comprise a fault tolerant computer system at the time of enforcing the data access method in a fault tolerant computer system. It is a figure which shows the module structure of the maintenance node which is contained in a fault tolerant computer system and performs processing, such as data acquisition from a processing node, presentation to a user, edit, overwrite update of the edited data to a processing node. It is a figure which shows the module structure of the processing node which is contained in a fault tolerant computer system and performs the process with respect to the request | requirement from the outside. The maintenance node that is included in the fault-tolerant computer system and performs processing such as data acquisition from the processing node, presentation to the user, editing, overwrite update of the edited data to the processing node, etc. It is a flowchart which shows the flow of the process displayed on a user. A maintenance node that is included in a fault-tolerant computer system and performs processing such as data acquisition from processing nodes, presentation to the user, editing, overwrite update of edited data to processing nodes, etc. It is a flowchart which shows the flow of the process which overwrites and updates. Data attribute information and access that is included in the fault-tolerant computer system and managed in the maintenance node that performs processing such as data acquisition from the processing node, presentation to the user, editing, overwriting and updating of edited data to the processing node It is a figure which shows the format of the profile information which defined authority, the presence or absence of possession in another processing node, the necessity of data content matching between processing nodes, etc. It is a flowchart which shows the flow of the process which performs data acquisition, the update monitoring of edit object data, and data overwrite update in the processing node which is contained in a fault tolerant computer system and performs the process with respect to the request | requirement from the outside. It is a figure which shows the example of a screen display when the data content corresponds between process nodes for showing the data acquired from the process node with respect to a user. It is a figure which shows the example of a screen display when there exists a difference in the data content between processing nodes for showing the data acquired from the processing node with respect to a user.

FIG. 1 is a diagram showing an outline of a data access method in a fault tolerant computer system.

Main components are a maintenance node 0211 that constitutes the fault-tolerant computer system 0201, and performs processing such as data acquisition from the processing node 0212, presentation to the user, editing, overwriting update of edited data to the processing node 0212, A plurality of processing nodes 0212 and a user terminal 0202 that execute processing in response to an external request in parallel. In response to an external request, the plurality of processing nodes 0212 execute the same processing at the same timing using the same data 0101 held in each node. In order to guarantee the operation of the fault-tolerant computer system 0201, the contents of data used by the application for processing must be always the same among a plurality of nodes executed in parallel. On the other hand, when the user refers to and edits the data 0101, accessing each processing node 0212 individually and editing the same data 0101 so as to have the same contents are burdens and cause human error. obtain. As a countermeasure against these problems, an outline of the method according to the present invention will be described below.

When the user refers to the same data 0101 held by a plurality of processing nodes 0212 via the user terminal 0202, the data is collected from each processing node 0212 to the maintenance node 0211 (0121). Here, in order to collect the data 0101, a copy of the corresponding data 0101 is generated at each processing node 0212, transmitted to the maintenance node 0211, and the data is stored on the maintenance node 0211 (0102). The maintenance node 0211 compares the contents of the collected data 0102 (0122), and displays the data contents on the screen 0111 of the user terminal 0202 based on the contents comparison result (0112). Here, if the contents of the data from all the processing nodes 0212 match as a result of the content comparison, it is designated in advance in the processing node 0212 which is a representative defined in advance or the processing node in operation. Only the content of data acquired from one processing node 0212 that matches the condition (matches all processing nodes) is displayed. If there is a difference in data content from at least one processing node 0212 as a result of the content comparison, the data content of a portion having a difference between the processing nodes 0212 is displayed.

When the user edits the data 0101 on the processing node 0212 via the user terminal 0202, one data is selected from the data 0102 collected from each processing node 0212 in 0121 (0103), and the data 0103 is stored in the data 0103. On the other hand, the user is allowed to edit using an editing tool. The edited data 0103 is distributed from the maintenance node 0211 to each processing node 0212, and the contents of the corresponding data 0101 in each processing node 0212 are overwritten and updated (0123).

From the above, it is possible to update the contents of the same data collectively to a plurality of processing nodes 0212 constituting the fault tolerant computer system 0201 only by referring to and editing one data as a user operation. In addition, there is no difference in data contents between the processing nodes 0212.

FIG. 2 is a diagram showing an overview of a fault-tolerant computer system that is configured by a plurality of independent nodes interconnected via a network (communication medium), and that the same processing is executed in parallel in each of the constituent nodes. It is.

The main components of the fault tolerant computer system 0201 are processing such as data acquisition from the processing node 0212, presentation to the user, editing, overwriting update of the edited data to the processing node 0212, which are interconnected via the LAN 0214. A maintenance node 0211 that performs the above, two or more processing nodes 0212, a gateway server 0213 that connects to the wide area network 0203 and relays communication with an external system, and the like. A user terminal 0202 connected to the maintenance node 0211 via the external network 0205 is used for information presentation for users, various operations, and the like.

The fault tolerant computer system 0201 receives a request from the external system 0204 that can communicate via the wide area network 0203, performs processing on the request, and returns a service result as a response to the external system 0204. provide. Here, the input message 0231 is received as a request from the external system 0204, and an output message 0241 storing the processing result for the request is transmitted to the external system 0204.

In the fault tolerant computer system 0201, the gateway server 0213 that has received the input message 0231 as a request from the external system 0204 via the wide area network 0203 sends an input message to all the processing nodes 0212 in the system via the LAN 0214. Transfer as 0232 again. Here, almost all the processing nodes 0212 receive the message and broadcast the input message 0232 so that the processing for the message can be started. Each processing node 0212 that has received the input message 0232 executes processing for the input message 0232 and transmits an output message 0242 storing the processing result to the gateway server 0213 via the LAN 0214. The gateway server 0213 that has received the output message 0242 from each processing node 0212 creates an output message 0241 as a response to the requesting external system 0204, and transmits the output message 0241 to the external system 0204. Here, the gateway server 0213 performs comparison and collation of data contents of one or more output messages 0242 received from each processing node 0212, correct / incorrect determination, etc., and sends correct message data to the requesting external system 0204 as an output message 0241. Send. In the comparison / collation and correctness / incorrectness determination of the data content of the output message 0242, the output message 0242 having the largest number of matching data contents is regarded as the correct message data, and one of the correct output messages 0242 is sent to the external system 0204. Output message 0241.

The maintenance node 0211 does not execute any online processing executed in response to the request from the external system 0204. When a user who logs in to the maintenance node 0211 using the user terminal 0202 refers to and edits data in the processing node 0212, the data is acquired and stored from each processing node 0212, and each processing node 0212 of the data after editing is stored. Transfer to, overwrite update, etc.

The main hardware configuration of the maintenance node 0211 includes a processing device (CPU) 0221, a storage device (memory, hard disk) 0222, and a communication device 0223. In the storage device 0222, a software program for acquiring data from the processing node 0212 and storing the data in a designated area in the storage device 0222, for comparing the contents of the data acquired from the processing node 0212 and presenting it to the user A software program for performing screen display, a software program for editing data acquired and stored from the processing node 0212, and data edited by the user for transferring to the processing node 0212 for overwriting update A software program, a software program for performing communication between the maintenance node 0211 and the processing node 0212 via the LAN 0214, and the like are stored, and these software programs are processed by the processing device 0221. The communication device 0223 performs communication processing when the processing node 0212 acquires data, communication processing for inputting / outputting a screen for logging in from the user terminal 0202, and the like.

Note that the processing device (CPU) 0221 reads a program necessary for processing from the storage device 0222 such as a hard disk and executes each processing. It also controls operations of the storage device 0222 and the communication device 0223. In addition, the processing of the maintenance node 0211 in the flowcharts and sequence diagrams described in the drawings to be described later is assumed to be executed by the processing device (CPU) 0221 unless otherwise specified.

The main hardware configuration of the processing node 0212 includes a processing device (CPU) 0224, a storage device (memory, hard disk) 0225, and a communication device 0226. The storage device 0225 stores data necessary for executing a service provided to the external system 0204 by the fault tolerant computer system 0201, and a user program for executing processing for a request from the external system 0204 regarding the service. , A software program for performing synchronization during operation among a plurality of processing nodes 0212, a software program for acquiring data in the processing node 0212 designated by the user, and transmitting the data to the maintenance node 0212, and a user terminal 0202 A software program for monitoring whether or not there is an update to the data on the processing node 0212 while the user is editing data at the maintenance node 0211 via the LAN, and the processing node 0212 and the gate via the LAN 0214 Software program for performing communication between the communication or processing node 0212 and maintenance node 0211 between Eisaba 0213 is stored, these software programs are processed by the processing device 0224. The communication device 0226 receives the input message 0232 from the gateway server 0213 and performs communication processing for transmitting the output message 0242 to the gateway server 0213.

Note that the processing device (CPU) 0224 reads a program necessary for processing from the storage device 0225 such as a hard disk, executes each processing, and controls operations of the storage device 0225 and the communication device 0226. In addition, the processing of the processing node 0212 in the flowcharts and sequence diagrams described in the drawings to be described later is assumed to be executed by the processing device (CPU) 0224 unless otherwise specified.

The main hardware configuration of the user terminal 0202 includes a processing device (CPU), a storage device (hard disk), and a communication device. The storage device stores a software program for logging in to the maintenance node 0211, performing command operations, screen display, and the like, a software program for performing communication between the user terminal 0202 and the maintenance node 0211, and the like. Processed by the processing device. The communication device performs communication processing for inputting / outputting a screen for login to the maintenance node 0211.

The processing device reads out a program necessary for processing from the storage device such as a hard disk and executes each processing. It also controls the operation of the storage device and the previous communication device. In addition, the processing of the user terminal 0202 in the flowcharts and sequence diagrams described in the drawings to be described later is executed by the processing device (CPU) unless otherwise specified.

FIG. 3 is a diagram showing an overall processing sequence between a plurality of nodes constituting the fault tolerant computer system when the data access method in the fault tolerant computer system is executed.

Main components are a user terminal 0202, a maintenance node 0211, a processing node 1 (0212a), a processing node 2 (0212b), and a processing node 3 (0212c) that constitute the fault-tolerant computer system 0201.

In 0301, the user terminal 0202 logs in to the maintenance node 0211 by a user operation. In 0302, data to be accessed by the user is designated by the user's operation. When the maintenance node 0211 accepts 0302 data designation, in 0311, when the user requests the corresponding data specified by the user to the processing node 1 (0212a), the processing node 1 (0212a) extracts the corresponding data in 0321, The corresponding data is transmitted to the maintenance node 0211. In 0312, when requesting the user-specified corresponding data to the processing node 2 (0212b), the processing node 2 (0212b) extracts the corresponding data in 0331 and transmits the corresponding data to the maintenance node 0211. In 0313, when requesting the corresponding data designated by the user to the processing node 3 (0212c), the processing node 3 (0212c) extracts the corresponding data in 0341 and transmits the corresponding data to the maintenance node 0211. Thereafter, the same processing as 0311 to 0313 is performed on all the processing nodes 0212 constituting the fault tolerant computer system 0201, and the corresponding data designated by the user is acquired from all the processing nodes 0212. In 0314, the data acquired in 0311 to 0313 is stored in a storage area (hard disk or the like) in the maintenance node 0211. In 0315, the contents of the data acquired from the processing node 0211 are compared in 0311 to 0313. In 0316, based on the content comparison result in 0315, the data content is displayed on the user terminal 0202. Details of the processing and display method in 0315 and 0316 will be described with reference to FIGS.

In 0303, the user terminal 0202 starts data editing for the maintenance node 0211 by a user operation. When the maintenance node 0211 accepts the start of data editing by the user 0303, the editing tool is activated at 0317 and the corresponding data to be edited is opened. Thereafter, data editing work by the user is performed. In 0304, data editing for the maintenance node 0211 is terminated by the user's operation. When the maintenance node 0211 accepts the completion of data editing by the user 0304, in 0318, the processing node 0212 (all processing nodes or individual processing nodes) to be updated with respect to the corresponding data edited by the user is determined. In 0319, the data edited by the user is distributed to the processing node 0212 to be updated determined in 0318. The processing node 0212 that has received the edited data distributed from the maintenance node 0211 performs overwrite update on the corresponding data (0322, 0332, 0342). Details of the processing in 0317 to 0319 will be described with reference to FIG. Details of the processes at 0322, 0332, and 0342 will be described with reference to FIG.
FIG. 4 is a diagram showing a module configuration of a maintenance node that is included in the fault tolerant computer system and performs processing such as data acquisition from the processing node, presentation to the user, editing, overwriting update of the edited data to the processing node, and the like. is there.

The maintenance node 0211 includes a data access management unit 0401 and a processing node 0212 that perform processing such as data acquisition from the processing node 0212, presentation to the user, management of editing by the user, and overwrite update of the edited data to the processing node 0212. A hard disk 0402 for storing the acquired data and an editing tool 0403 for the user to edit the data are introduced. The data access management unit 0401 is a software program executed by the processing device (CPU) 0221.

The main component of the data access management 0401 is data acquisition for acquiring data from the processing node 0212 via the data communication unit 0417 based on the designation from the user via the screen display unit 0415 and storing the data in the hard disk 0402. A storage unit 0411, a data content comparison unit 0412 that compares data contents acquired by the data acquisition / storage unit 0411 from a plurality of processing nodes 0212, and an editing tool based on designation from the user via the screen display unit 0415 A data editing management unit 0413 that monitors activation and termination of 0403, a data distribution / update unit 0414 that distributes data edited by the user using the editing tool 0403 to the processing node 0212 and overwrites and updates the corresponding data, and communication Remote connection from user terminal 0202 via medium 0205 A screen display unit 0415 that provides a command input screen, an output result output screen of the data content comparison unit 0412, a screen of the editing tool 0403, and the like, and a screen for accepting a remote connection from the user terminal 0202 via the communication medium 0205 A remote access unit 0416 connected to the display unit 0415, and a data communication unit 0417 for performing communication between the processing node 0212 and the like in the system via the communication medium 0214. The profile information 0421 defines, for each data held by the processing node 0212, data attribute information, access authority, presence / absence of possession at other processing nodes, necessity / unnecessity of data content matching between processing nodes, and the like. The data content comparison unit 0412, the data distribution / update unit 0414, and the like refer to these in each process. Details of the profile information 0421 will be described in FIG.

FIG. 5 is a diagram showing a module configuration of a processing node included in the fault-tolerant computer system and executing processing for an external request.

The processing node 0212 includes a data access management unit 0501 that performs processing such as acquisition of designated data and transmission to the maintenance node 0211, a user program 0502 that performs various processing in response to requests from the external system 0204, and middleware 0503. In addition, a hard disk 0504 and a memory 0505, which are storage destinations of data to be referenced and updated by the user program 0502, middleware 0503, are introduced. The data access management unit 0501 is a software program executed by the processing device (CPU) 0224.

The main component of the data access management unit 0501 receives a data acquisition request from the maintenance node 0211 via the data communication unit 0513, extracts the relevant data from the hard disk 0504 or the memory 0505, and sends the relevant data to the maintenance node 0211. Data management unit 0501 that receives a data overwrite update request from the maintenance node 0211 and performs overwrite update on the corresponding data, and processing to the original data transmitted by the data management unit 0501 to the maintenance node 0211 A data update monitoring unit 0512 that monitors updates by a program or the like in the node 0211 and a data communication unit 0513 that performs communication with the maintenance node 0211 or the like in the system via the communication medium 0214.

FIG. 6 is a maintenance node that is included in the fault tolerant computer system and performs processing such as data acquisition from the processing node, presentation to the user, editing, overwriting update of the edited data to the processing node, and the like from the processing node. It is a flowchart which shows the flow of the process which displays the acquired data to a user.

In 0601, a data path of data to be accessed is acquired from an operation by the user via the user terminal 0202. In 0602, data corresponding to the data path acquired in 0601 is requested to the processing node 0212 and acquired. In 0603, the data acquired from the processing node 0212 in 0602 is stored in a storage area (such as a hard disk) in the maintenance node 0211. In 0604, if data acquisition from all the processing nodes 0212 has not been completed yet, the processing of 0602 and 0603 is repeated. If the data acquisition from all the processing nodes 0212 has been completed in 0604, the contents of the same data acquired from the processing nodes 0212 in 0602 to 0604 are compared in 0605.
Here, a general tool such as “diff” may be used to compare the data contents. In 0606, if the data contents from all the processing nodes 0212 match as a result of the data content comparison in 0605, the profile information of the corresponding data is referred to in 0607. In 0608, when the “content match” section of the profile information of the corresponding data indicates that the contents should not match between the processing nodes 0212 (NO), in 0610, a warning to the user (data contents match) Although it is not allowed to be displayed, it is displayed that it matches. In 0609, the contents of data acquired from one processing node 0212 that matches the condition specified in advance in the representative processing node 0212 defined in advance or the processing node in operation (matches with all processing nodes). indicate. In 0606, if there is a difference in the data contents from at least one processing node 0212 as a result of the data content comparison in 0605, the profile information of the corresponding data is referred to in 0611. In 0612, if the item “content match” in the profile information of the corresponding data indicates that the content needs to match between the processing nodes 0212 (YES), a warning to the user (data content matches) in 0613. Displayed to indicate that they do not match even though they are necessary. In 0614, the data contents of a portion having a difference between the processing nodes 0212 are displayed. Details of the display method here are shown in FIG. 10B.

FIG. 7 is a fault-tolerant computer system that is edited by a user at a maintenance node that performs processing such as data acquisition from a processing node, presentation to a user, editing, and overwrite update of edited data to a processing node. 6 is a flowchart showing a flow of processing for overwriting and updating updated data in a processing node.

In 0701, an instruction to start editing of data is received from an operation by the user via the user terminal 0202. In 0702, if the contents of the data from all the processing nodes 0212 match as a result of comparing the contents of the data acquired from the processing node 0212 as shown in FIG. 6, the representative defined in advance in 0703 Data acquired from one processing node 0212 that matches a condition specified in advance in the processing node 0212 or an active processing node is determined as data to be edited. In 0702, if there is a difference in the data contents from at least one processing node 0212 as a result of comparing the data contents acquired from the processing node 0212 as shown in FIG. 6, the processing node 0212 selected by the user in 0704 Is determined as data to be edited (here, a screen display for allowing the user to select a processing node as a data acquisition source is performed). In 0705, the editing tool is activated and the data determined in 0703 or 0704 is opened. Here, the editing tool is defined in advance for each type of data to be edited. For example, if it is a text file etc., it is a text editor etc. In 0706, the end of the editing tool started in 0705 is monitored. If the end is not detected, update of the corresponding data is detected on the processing node 0212 side in 0707, and the updated data is transmitted to the maintenance node 0211. If it is, in 0708, the user is notified of the occurrence of the update, the processing node 0212 is presented with a selection as to whether or not to replace it with the updated data, and the data replacement is executed or not executed according to the user's selection. I do. Thereafter, the processing returns to 0706. In 0707, if the processing node 0212 has not detected the update of the corresponding data, the processing returns to 0706. In 0706, the end of the editing tool activated in 0705 is monitored. If the end is detected, in 0709, the user is made to select the processing node 0212 that is the overwrite update destination of the edited data. Here, it is selected whether to perform overwrite update collectively for all the processing nodes 0212, or to perform overwrite update by individually specifying one processing node. When the user selects overwrite update to all processing nodes, the profile information of the corresponding data is referred to in 0710. In 0711, if the “content match” section of the profile information of the corresponding data indicates that the content needs to be matched between the processing nodes 0212 (YES), then in 0712, the user is sent to all the processing nodes 0212. Update the edited data by overwriting. In 0711, if the “content match” section of the profile information of the corresponding data indicates that the content should not match between the processing nodes 0212 (NO), in 0713, a warning is given to the user (all processing nodes Is not possible), and overwrite update is not performed. In 0709, when the user selects overwrite update in which one processing node is individually specified, in 0714, the profile information of the corresponding data is referred to. In 0715, if the “content match” section of the profile information of the corresponding data indicates that the content needs to match between the processing nodes 0212 (YES), a warning (all processing nodes) is issued in 0716. (The content must match in the above) is displayed, and overwriting is not performed. In 0715, if the “content match” section of the profile information of the corresponding data indicates that the content should not match between the processing nodes 0212 (NO), in 0717, for each processing node 0212, The user overwrites and edits the edited data. In 0718, the execution result of the overwrite update is received from each processing node 0212. If all the processing nodes 0212 that have performed the overwrite update succeed, in 0719, an end instruction is transmitted to all the processing nodes 0212 that have performed the overwrite update, and the processing ends. In 0718, if at least one of the processing nodes 0212 that have performed the overwrite update fails, in 0720, an instruction to restore the data is sent to all the processing nodes 0212 that have performed the overwrite update. ,finish.

FIG. 8 is a diagram of a fault-tolerant computer system that manages data in a maintenance node that performs processing such as data acquisition from a processing node, presentation to a user, editing, overwriting and updating of edited data in a processing node. It is a figure which shows the format of the profile information which defined attribute information, access authority, the presence or absence of possession in another processing node, the necessity of data content matching between processing nodes, etc.

Main components of profile information are a type 0801, a data path 0802, an access authority 0803, an access content 0804, another node possession 0805, and a content match 0806. Setting values corresponding to these items are specified for each data.

The type 0801 describes the type of data. The data path 0802 describes information for specifying the location of data. For example, if the type 0801 is a file, the file path is described, and if the type 0801 is memory data, the address and size are described. The access authority 0803 describes the authority of the user who is permitted to access the corresponding data. The access content 0804 describes the access content that can be executed by the user permitted by the access authority 0803. In the example of the figure, “R” (reading is possible), “W” (writing is possible), and the like are described. In other node possession 0805, whether or not the relevant data is possessed by other processing node 0212 is described (YES / NO). The content match 0806 describes whether the data content needs to be matched between the processing nodes 0212 or not (YES / NO) when the corresponding data is held in other processing nodes 0212. To do.

The profile information is created for each processing node 0212 in the form of a file, etc., and is stored and managed by the maintenance node 0211. The content data of the same data acquired from a plurality of processing nodes 0212 and the data edited by the user are displayed. It is referred to when executing processing such as overwriting update for a plurality of processing nodes 0212. The profile information is created in advance by the administrator or user of the fault tolerant computer system and is updated as appropriate.

FIG. 9 is a flowchart showing the flow of processing that is included in the fault-tolerant computer system and performs data acquisition, update monitoring of data to be edited, data overwrite update, and the like in a processing node that executes processing in response to an external request. is there.

In 0901, a data acquisition request is received from the maintenance node 0211. In 0902, data corresponding to the request in 0901 is acquired in the processing node 0212 and transmitted to the maintenance node 0211. In 0903, the corresponding data acquired in 0902 is duplicated, and the duplicated data is stored in a temporary storage area (a temporary use directory in the hard disk, etc.). In 0904, monitoring of update by the program in the processing node 0212 or monitoring of update of the corresponding data by another user is started for the data requested by the maintenance node 0211. In 0905, when an update is detected as a result of the update monitoring in 0904, the corresponding data after the update is transmitted to the maintenance node 0211 together with an update notification in 0906. Thereafter, the processing returns to 0905. In 0905, if no update has been detected as a result of the update monitoring in 0904, in 0907, if an instruction to overwrite the corresponding data from the maintenance node 0211 and edited data by the user are received, in 0908, the corresponding data is processed. The user overwrites and updates the user-edited data received from the maintenance node 0211. Thereafter, the processing returns to 0905. In 0909, when an instruction to restore the corresponding data from the maintenance node 0211 is received, in 0910, the corresponding data is copied data (data before editing / updating) stored in the temporary storage area in 0903. Overwrite and update. In 0912, the duplicated data (data before editing / updating) is deleted from the temporary storage area. In 0911, when an end instruction from the maintenance node (overwrite update of the corresponding data to the processing node 0212 is normally completed) is received, the duplicated data (data before editing / updating) is deleted from the temporary storage area in 0912 To do.

FIG. 10A is a diagram showing a screen display example when the data contents match between the processing nodes for presenting the data acquired from the processing nodes to the user.

The screen display 1001 is displayed when the contents of the data from all the processing nodes 0212 match as a result of comparing the contents of the same data acquired from the plurality of processing nodes 0212 shown in FIG. The data on the processing node 0212 is displayed as an output result for the command operation of the designated user. In 1011, attribute information (type and name) of the acquired data and a content comparison result (data from all processing nodes 0212 match in the example in the figure) are displayed. In 1012, the contents of the acquired data are displayed as they are.

FIG. 10B is a diagram illustrating a screen display example when there is a difference in data contents between processing nodes for presenting data acquired from the processing nodes to a user.

The screen display 1002 shows a case where there is a difference in data contents from at least one processing node 0212 as a result of comparing the contents for the same data acquired from a plurality of processing nodes 0212 shown in FIG. The data on the processing node 0212 is displayed as an output result for the command operation of the designated user. In 1021, attribute information (type and name) of the acquired data and a content comparison result (in the example of the figure, there is a difference in data from one processing node 0212) are displayed. In 1021, the data content of a portion having a difference between the one processing node 0212 having the difference and the other processing node 0212 is displayed in a comparison format.

The data access method in the fault-tolerant computer system according to the present invention can be used particularly when the user performs work such as debugging and testing of an application program running on the fault-tolerant computer system 0201.

For example, when editing a definition file of an application program for testing and debugging and changing a setting value or the like, the method according to the prior art accesses one processing node 0212 at a time and accesses the same file with the same contents. It was necessary to carry out updates. In addition, it is necessary for the user to check whether the contents of the corresponding files in all the processing nodes 0212 are the same. If there is a difference in the contents of the corresponding file between the processing nodes 0212, the application program may operate differently between the processing nodes 0212, and the fault-tolerant computer system 0201 becomes in an abnormal state, and the test and debugging may not be performed correctly. is there. On the other hand, if the method according to the present invention is used, the user only has to open the file once and update the contents. Are guaranteed to match. Furthermore, updating of node-specific information such as an IP address is restricted so that all processing nodes 0212 do not have the same content. Therefore, even if the user updates the file, the application program does not operate differently between the processing nodes 0212 as described above.

Also, when performing failure analysis or the like by referring to a log of an application or the like, in the conventional method, it is necessary to access each processing node 0212 and open and refer to the corresponding log file. Further, it is necessary to compare and refer to log information for the same processing described in the log file between the processing nodes 0212 by the user himself and find a difference. For this reason, when the amount of logs increases, the load on the user also increases. On the other hand, in the method according to the present invention, the user only needs to open the corresponding file once, and the location having a difference between the processing nodes 0212 is automatically displayed. Become.

In addition, for example, when definition information, reference data, etc. are transferred from an external system or the like by remote operation to the fault tolerant computer system 0201 and stored in each processing node 0212, human error, network failure, waiting, etc. May cause processing nodes 0212 that fail to store data, refer to the corresponding data using the method of the present invention, and whether the definition information, reference data, etc. are correctly stored in each processing node 0212 Confirmation is effective because the workload is low.

As mentioned above, although embodiment of this invention was described concretely based on the embodiment, it is not limited to this and can be variously changed in the range which does not deviate from the summary.

0101 Data 0111 User terminal screen 0201 Fault tolerant computer system 0202 User terminal 0211 Maintenance node 0212 Processing node

Claims

A method in a fault tolerant computer system comprising a plurality of processing nodes connected via a network and a maintenance node for acquiring, editing, and updating data in the plurality of processing nodes, wherein the plurality of processing nodes have the same data When the user performs the same process in parallel using the same and the user refers to or edits the same data of the plurality of processing nodes, the maintenance node processes the same data in the plurality of processing nodes. The step of acquiring and storing from the node via the network, and the maintenance node comparing the content of the same data acquired from the plurality of processing nodes, and displaying the content to the user via the user terminal based on the comparison result Steps to perform and pre-defined representative processing nodes or running A step in which a user edits on a maintenance node via a user terminal for data acquired from one processing node that matches a condition specified in advance in the processing node; Transferring the data to the plurality of processing nodes and overwriting and updating the data.
The method according to claim 1, wherein the fault tolerant computer system receives a processing request from an external device via a network, or transmits a processing content for the processing request to the external device via a network. A gateway device, the gateway device receives a processing request received from the external device via a network;
The processing request is transmitted to the plurality of processing nodes, the plurality of processing nodes respectively execute processing for the processing request in parallel, and the executed processing content is transmitted to the gateway device, respectively, and the gateway device Is a method of collating processing contents executed at each processing node received from the plurality of processing nodes, and transmitting the processing contents regarded as normal to the external apparatus via the network.
The method according to claim 1, wherein the maintenance node has, for each piece of data held by the plurality of processing nodes, data attribute information, access authority, presence / absence of holding at other processing nodes, between processing nodes. A method of holding profile information that defines whether or not data content match is necessary.
The method according to claim 1 or 3, wherein the maintenance node compares contents of the same data acquired from the plurality of processing nodes, and compares the contents when the contents are displayed to the user via the user terminal. As a result, when the data contents of all of the plurality of processing nodes match, only the data of one representative processing node is displayed, the profile information of the data is referred to, and if the data does not match, the user If there is a difference in the data contents of at least one processing node as a result of the comparison, the contents of the difference parts are displayed side by side. A method of warning the user if it is data.
The method according to claim 1 or 3, wherein when the user edits the same data acquired and stored from the plurality of processing nodes via the user terminal by the maintenance node based on the content comparison result. When the data contents of all of the plurality of processing nodes match, the editing process is performed on the data acquired and stored from one representative processing node, and the data contents of at least one processing node are different. If there is, a method of selecting from which processing node to edit the data acquired and saved.
The method according to claim 1 or 3, wherein when the maintenance node transfers data edited by a user to the plurality of processing nodes and overwrites and updates it, the profile information of the data is referred to, A method comprising determining whether to perform data transfer and overwrite update for all processing nodes, or to perform data transfer and overwrite update for only one corresponding processing node.
4. The method according to claim 1 or 3, wherein when the maintenance node transfers the data edited by the user to the plurality of processing nodes and overwrites and updates the data, all of the plurality of processing nodes are selected by the user. A method, comprising: determining whether to perform data transfer and overwrite update for a corresponding processing node or to perform data transfer and overwrite update for only one corresponding processing node.
8. The method according to claim 1, wherein when the maintenance node performs data transfer and overwrite update for all of the plurality of processing nodes, the data before all of the plurality of processing nodes is updated. Is recorded, and when data transfer to at least one processing node and overwriting update fails, the data of all of the plurality of processing nodes is returned to the content before update.
6. The method according to claim 1 or 5, wherein when a user edits data on the maintenance node via a user terminal, whether the data is updated by a program on the processing node, other users The presence or absence of the update of the data is monitored, and when the update is detected, the user is notified, and whether or not the content of the data is replaced with the updated content is determined by the user's selection how to.
A fault tolerant computer system comprising a plurality of processing nodes that execute the same processing in parallel using the same data connected via a network, and a maintenance node that acquires, edits, and updates data in the plurality of processing nodes Because
The maintenance node is
A storage device for storing the same data acquired from each processing node via a network;
Compares the contents of the same data acquired from the multiple processing nodes, displays the comparison result on the user terminal, and specifies in advance among the processing nodes that are defined in advance or are in operation A processing device for displaying acquired data from one processing node that matches a specified condition on a user terminal, storing an editing result by a user in the storage device, and transferring the edited data to the plurality of processing nodes When,
A fault tolerant computer system comprising: