WO2010106593A1 - 高信頼性計算機システムおよびその構成方法 - Google Patents
高信頼性計算機システムおよびその構成方法 Download PDFInfo
- Publication number
- WO2010106593A1 WO2010106593A1 PCT/JP2009/005872 JP2009005872W WO2010106593A1 WO 2010106593 A1 WO2010106593 A1 WO 2010106593A1 JP 2009005872 W JP2009005872 W JP 2009005872W WO 2010106593 A1 WO2010106593 A1 WO 2010106593A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- application
- computer
- processing unit
- storage device
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/2097—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements maintaining the standby controller/processing unit updated
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/202—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant
- G06F11/2038—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant with a single idle spare processing component
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/202—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant
- G06F11/2046—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where processing functionality is redundant where the redundant components share persistent storage
Definitions
- the present invention relates to a technique for configuring a highly reliable computer system for applications that require high reliability.
- clustering As one of the techniques for realizing such high availability, a system configuration technique called clustering is known in which an active system and a standby system are prepared, and switching from the active system to the standby system occurs when a problem occurs in the active system. .
- Clustering methods include (a) a method that prevents the active system from holding the processing state, and only switches between the active system and the standby system when a failure in the active system is detected, or (b) the active system and the standby system.
- a method is known in which the states of these are matched and the processing at the time of failure detection is recovered when a failure is detected. Since it is difficult for the method (a) to give the operating system a state, the method (b) has higher applicability.
- the I / O state is buffered in the operation system until the synchronization of the memory state is completed, and the I / O state is reflected when the synchronization between the systems is completed.
- Is a restart point and when a failure is detected, a technique for re-executing the standby system from the restart point is shown (see Non-Patent Document 1).
- hypervisor virtualizes the entire system (system virtualization) of a hardware system that executes an application and an OS.
- the present invention has been made in view of the above-described problems of the prior art, and an object of the present invention is to provide a highly reliable computer system capable of speeding up the copying process and a configuration method thereof.
- the present invention monitors the state of the program of the active computer, detects a synchronization point for performing state synchronization between the active computer and the standby computer, Only information necessary for continuation of subsequent processing is extracted from the storage device of the active computer as information to be copied, and the extracted copy target information is copied from the active computer to the standby computer. Is.
- (A) is a state diagram which shows the state of the memory under application execution
- (b) is a state diagram which shows the state of the memory at the time of application termination.
- (A) is a state diagram showing the state of the memory at the end of the processing phase # 1
- (b) is a state diagram showing the state of the memory at the end of the processing phase # 2. It is a flowchart for demonstrating the synchronization point determination process in the case of setting the time of process phase switching as a synchronization point.
- (A) is a state diagram which shows the state of the memory before GC completion
- (b) is a state diagram which shows the state of the memory after GC completion.
- an application the end point of an application program (hereinafter referred to as an application) is used as a synchronization point, and information in unnecessary areas (unused areas) is not copied.
- FIG. 1 is a configuration diagram of a highly reliable computer system showing a first embodiment of the present invention.
- the high-reliability computer system is composed of an active computer 101 and a standby computer 102.
- the active computer 101 and the standby computer 102 are connected by a connection network 103 such as a network or a bus. Via the shared external storage device 120.
- the operational computer 101 includes hardware 104 as operational hardware resources, and includes a system virtualization processor 105, an application execution OS (Operating System) 106, and an application virtualization processor 107 as operational software resources. And an application 108 and a management OS 109.
- a system virtualization processor 105 includes a system virtualization processor 105, an application execution OS (Operating System) 106, and an application virtualization processor 107 as operational software resources.
- an application 108 and a management OS 109 includes a management OS 109.
- the standby computer 102 has basically the same configuration as that of the active computer 101.
- the standby computer 102 includes hardware 114 as a standby hardware resource, and includes a system virtualization processing unit 115 and an application execution resource as a standby software resource.
- the system includes an OS 116, an application virtualization processing unit 117, an application 118, and a management OS 119.
- the hardware 104, 114 includes, for example, an input / output device, a storage device (hereinafter referred to as a memory), a processing device (both not shown), and the like.
- a memory stores a plurality of programs including a control program and a processing program, and information constituting each software resource.
- the system virtualization processing unit 105 performs processing by virtualizing the hardware 104 for the application execution OS (Operating System) 106, the application virtualization processing unit 107, the application 108, and the management OS 109.
- the processing unit 107 virtualizes the application 108 with respect to the application execution OS 106 and executes processing.
- the system virtualization processing unit 105 monitors the execution state of the application execution OS and the application 108 to detect a synchronization point for state synchronization with the standby computer 102, and at the detected synchronization point, the memory Information to be copied necessary for continuation of processing is extracted, and the extracted information to be copied is transferred to the standby computer 102 via the connection network 103.
- the system virtualization processing unit 105 includes a state copy processing unit 110 that is a characteristic process of the present invention.
- the state copy processing unit 110 extracts state information related to the state of the memory used by the OS 106 operating on the system virtualization processing unit 105 and the application virtualization processing unit 107 and the application 108 from the memory as information to be copied.
- the extracted status information is transferred to the standby computer 102 via the connection network 103, and the standby computer 102 is instructed to copy the status information.
- the operational computer 101 sends the I / O operation issued from the OS 106 to the system virtualization processing unit 105 to the management OS 109 once, and the management OS 109 performs buffering of the I / O operation.
- the buffer 201 stores data associated with buffering.
- the I / O operation buffered by the management OS 109 is reflected from the buffer 201 to the hardware 104 by the system virtualization unit 105 when the copying of the state information from the active computer 101 to the standby computer 102 is completed. .
- the I / O operations reflected in the hardware 104 are similarly buffered by the active computer 101 and the standby computer 102 by sending external input information to the active computer 101 and the standby computer 102.
- FIG. 3 shows a processing sequence 301 of the active computer 101 and a processing sequence 302 of the standby computer 102.
- the active computer 101 detects the synchronization point 303
- the active computer 101 copies the memory status information 304 to the standby computer 102 at this synchronization point 303.
- the active computer 101 buffers I / O operations after the synchronization point 303 (305).
- the standby computer 102 resumes processing from the start point 307 based on the copied state information 304. Will be.
- the standby computer 102 continues the processing in the active computer 101.
- FIGS. 4A and 4B show the state of the memory accompanying the execution of the application 108.
- FIG. FIG. 4A shows the state of the memory during execution of the application 108.
- the storage area 400 of the memory includes a use area 401 of the OS 106, a use area 402 of the first application (AP # 1), a use area 403 of the second application (AP # 2), and an unused area. 404.
- FIG. 4B shows a state where the execution of the first application (AP # 1) has been completed (completed).
- the memory storage area 400 includes a use area 401 of the OS 106, an execution end area 405, a use area 403 of the second application (AP # 2), and an unused area 404.
- the execution end area 405 is an area corresponding to the use area 402 used by the first application (AP # 1), and is regarded as an unused area.
- the use of the OS 106 is performed.
- Information about the area 401, the use area 402 of the first application (AP # 1), the use area 403 and the unused area 404 of the second application (AP # 2) is all from the active computer 101 to the standby computer 102. Will be copied.
- the end point of the application 108 is set as a synchronization point, and copying of unnecessary area (unused area) information is not performed, thereby speeding up the state information copying process.
- the processing shown in FIG. 5 is executed by the state copy processing unit 110 in the system virtualization processing unit 105.
- the processing by the state copy processing unit 110 is activated in response to an appropriate factor in the process of realizing system virtualization.
- the state copy processing unit 110 starts processing in step 501, and then investigates the operations of the OS 106, the application virtualization processing unit 107, and the application 108 that are operating on the system virtualization processing unit 105. Whether or not the synchronization point is reached is determined based on the execution state of step 108 (step 502). When the execution of the application 108 is finished, the process proceeds to step 503. When the execution of the application is not finished, step The process proceeds to step 509, and the process in this routine ends.
- step 502 The specific processing content of step 502 is shown in FIG.
- the state copy processing unit 110 executes determination of synchronization points and calculation of a non-target region set.
- the state copy processing unit 110 starts processing in step 601 and then determines whether or not the application 108 has ended (step 602). If it is determined in step 602 that the application 108 has ended, the state copy processing unit 110 sets the determination value S to “1”, for example, as a synchronization point, and ends the execution of the application for the non-target area N The execution end area is set (step 603), and then the process proceeds to step 605 to complete the processing in this routine.
- the execution of the first application (AP # 1) of the applications 108 is completed and the memory storage area 400 is configured as shown in FIG. 4B, the application (AP # 1) is used.
- the execution end area 405 corresponding to the used area 402 is removed from the copy target and is set as the non-target area N.
- step 602 determines whether the application 108 has ended. If it is determined in step 602 that the application 108 has not ended, the state copy processing unit 110 sets the determination value S to “0”, for example, as an asynchronous point (step 604). The process proceeds to 605 and the processing in this routine is completed.
- the state copy processing unit 110 is configured such that the storage area 400 of the memory is as shown in FIG. 4A, and the first application (AP # 1) and the second application (AP # 2) are included. When in the execution state, it is determined that it is not a synchronization point.
- step 503 the state copy processing unit 110 sets a variable N to a variable R, which is a set of areas used by the OS 106, the application virtualization processing unit 107, and the application 108 operating on the system virtualization processing unit 105. Find a region set that is not to be copied.
- the memory storage area 400 has four areas. Since it is divided into (the use area 401 of the OS 106, the execution end area 405, the use area 403 of the second application (AP # 2), and the unused area 404), the variable R of the area set is obtained as 4. And the variable N of the non-target region set is obtained as 2. In this case, the non-target area set includes an execution end area 405 and an unused area 404.
- the state copy processing unit 110 determines whether or not the region set variable R is an empty set (step 504). If the region set variable R is not an empty set, the process proceeds to step 505. Take one element. Subsequently, the state copy processing unit 110 determines whether or not the variable r is included in the variable N of the non-target region set (step 506). If the variable r is included in the variable N of the non-target region set, The control unit of the secondary storage device returns to step 504 and repeats the processing from step 504 to step 506 until the variable R of the area set becomes an empty set.
- step 506 If it is determined in step 506 that the variable r is not included in the variable N of the non-target region set, the state copy processing unit 110 proceeds to step 507 and determines a region outside the non-target region, that is, a copy target. In order to copy information stored in the usage area 401 of the OS 106 and the usage area 403 of the second application (AP # 2) from the active computer 101 to the standby computer 102 as information to be copied. Execute the process.
- step 504 If it is determined in step 504 that the variable R of the area set is an empty set, the state copy processing unit 110 assumes that all information to be copied has been copied from the active computer 101 to the standby computer 102, and step 508 is performed. Then, the buffered I / O operation is reflected on the hardware 104, the process proceeds to step 509, and the process in this routine is terminated.
- the synchronization point is set, and the use area 401 of the OS 106 is stored in the storage area 400 of the memory at this synchronization point.
- the information stored in the use area 403 of the second application (AP # 2) (information belonging to the application program scheduled to be used after the synchronization point), and the extracted information is a copy target necessary for continuing processing Since the information is copied from the active computer 101 to the standby computer 102, it is possible to speed up the copying process of information necessary for continuation of processing, which contributes to the improvement of the execution performance of the highly reliable computer system. be able to.
- the synchronization point is used.
- the execution of the first application (AP # 1) of the applications 108 is completed.
- the execution of the second application (AP # 2) is completed, the synchronization point is used. It can also be. In this case, only the information stored in the use area 401 of the OS 106 is copied from the active computer 101 to the standby computer 102 as information to be copied that is necessary for continuing the processing.
- a switching point of processing phases constituting the application 108 is set as a synchronization point, and the other configuration is the same as that of the first embodiment.
- FIG. 7B shows the state of the memory in phase # 2.
- the memory storage area 400 shown in FIG. 7A includes an OS use area 411, a use area 412 and an unused area 413 for the application 108.
- the usage area 412 of the application 108 includes application usage areas 414, 415, and 416 that are used only in the first processing phase # 1. For this reason, when the program shifts to the second processing phase # 2, the application use areas 414, 415, and 416 in the first processing phase # 1 are executed end areas 417, 418, 419, and the use area 412 of the application 108 becomes the application use area 420.
- the OS use area 411, the application usage area 412 or 420, and the information regarding the unused area 413 are all copied from the active computer 101 to the standby computer 102.
- the process phase switching point is used as a synchronization point, and information of unnecessary areas (unused areas 413, execution end areas 417, 418, and 419) is not copied, thereby copying state information. Is going to speed up.
- the processing in the present embodiment is the same as that in the first embodiment except for the synchronization point determination and the non-target region N setting. In this embodiment, the synchronization point determination and the non-target region N setting processing are performed. Only will be described. Further, the processing shown in FIG. 8 is executed by the state copy processing unit 110 in the system virtualization processing unit 105.
- the state copy processing unit 110 starts processing in step 801, and then monitors the execution state of the application 108 to determine whether or not the processing phase has ended (step 802). For example, when it is determined in step 802 that the processing phase # 1 has ended, the state copy processing unit 110 sets the determination value S to “1”, for example, as a synchronization point, and processes the non-target region N. The execution of the phase is set as the execution end region (step 803), and then the process proceeds to step 805 to complete the processing in this routine.
- the state copy processing unit 110 excludes the non-target area N excluded from the copy target from the use area 412 of the old process phase (process phase # 1) and the use area 420 of the new process phase (process phase # 2).
- the process of (execution end areas 417, 418, 419, unused area 413) is executed.
- step 802 determines whether the processing phase has ended. If it is determined in step 802 that the processing phase has ended, the state copy processing unit 110 sets the determination value S to, for example, “0”, assuming that it is an asynchronous point (step 8004), and then step The process proceeds to 805 to complete the processing in this routine.
- the processing phase switching point in the application 108 at which the execution of the first processing phase # 1 is completed is set as a synchronization point, and the OS 106 stores the memory 106 in the memory area 400 at this synchronization point. Only the information stored in the use area 411 and the area obtained by removing the execution end areas 417, 418, and 419 from the application use area 420 (information belonging to the processing phase scheduled to be used after the synchronization point) is extracted. Since the information to be copied necessary for continuation of processing is copied from the active computer 101 to the standby computer 102, the copying process of information necessary for continuation of processing can be speeded up, and a highly reliable computer system It is possible to contribute to the improvement of execution performance.
- the processing phase switching point at which the execution of the first processing phase # 1 in the application 108 is completed is set as the synchronization point, but the processing phase switching point at which the execution of the other processing phases is completed. Can also be used as a synchronization point. In this case, only information belonging to the processing phase scheduled to be used after the synchronization point is copied from the active computer 101 to the standby computer 102 as information to be copied that is necessary for continuing the processing.
- the point in time when the unused area of the application 108 is determined is set as a synchronization point, and the other configuration is the same as that of the first embodiment.
- the application virtualization processing unit 107 is an execution system provided with garbage collection (GC)
- the time when the unused area is determined by garbage collection (GC) is set as the synchronization point. Is.
- FIG. 9A shows the state of the memory before garbage collection (GC)
- FIG. 9B shows the state of the memory after garbage collection (GC).
- the memory storage area 400 shown in FIG. 9A includes an OS use area 421, an application use area 421, and an unused area 423.
- a plurality of unused data areas 424 exist in the application usage area 421 in a distributed manner.
- the OS use area 421 and all information regarding the application use area 421 and the unused area 423 are copied from the active computer 101 to the standby computer 102.
- the time when an unused area is determined by garbage collection (GC) is set as a synchronization point, and information of unnecessary areas (unused area 423 and a plurality of unused data areas 424) is not copied. Therefore, the state information copying process is accelerated.
- GC garbage collection
- the state copy processing unit 110 starts processing in step 1001, instructs the application virtualization processing unit 107 to execute garbage collection (GC), and determines whether or not garbage collection (GC) has been completed ( Step 1002).
- GC garbage collection
- the application virtualization processing unit 107 uses garbage collection (GC) to collect information on a plurality of unused data areas 424 belonging to the application virtualization usage area 421, and displays the collected information in FIG. ),
- the application virtualization usage area 425 is stored in the unused data area 426, and the application virtualization usage area 425 stores the unused data area 426 for storing unused data and the data in use.
- the status copy processing unit 110 is informed that the process is divided into the in-use data area 427 and configured. Notice.
- the state copy processing unit 110 When the status copy processing unit 110 receives a notification from the application virtualization processing unit 107 that the unused data area 426 has been determined, the state copy processing unit 110 indicates a synchronization point indicating when the unused area has been determined by the completion of garbage collection (GC). For example, the determination value S is set to “1”, and the non-target area N is set as an unused area determined by the completion of garbage collection (GC) (step 1003). Thereafter, the process proceeds to step 1005. The processing in this routine is completed.
- GC completion of garbage collection
- the unused data area 426 of the application virtualization usage area 425 is It is removed from the copy target area and is designated as a non-target area N.
- the state copy processing unit 110 uses information stored in the use area 421 and the in-use data area 427 of the OS 106, which is an area different from the non-target area N, that is, an area to be copied, as an operational computer. Processing for copying from 101 to the standby computer 102 is executed.
- step 1002 determines whether the garbage collection (GC) has been completed. If it is determined in step 1002 that the garbage collection (GC) has not been completed, the state copy processing unit 110 sets the determination value S to, for example, “0”, assuming that it is an asynchronous point (step 1004). Thereafter, the process proceeds to step 805 to complete the processing in this routine.
- GC garbage collection
- a point in time when an unused area is determined by the completion of garbage collection (GC) is set as a synchronization point, and at this synchronization point, the use area 421 of the OS 106 is stored as information stored in the storage area 400 of the memory. And only the information stored in the in-use data area 427 in the application virtualization utilization area 425 is extracted, and the extracted information is used as information to be copied necessary for continuing processing from the active computer 101 to the standby system. Since copying to the computer 102 is performed, it is possible to speed up the copying process of information necessary for continuation of processing, and to contribute to improving the execution performance of the highly reliable computer system.
- GC garbage collection
- the OS 106 operating on the system virtualization processing unit 105, the application virtualization processing unit 107, or an API (Application Programming Interface) call from the application 108 is used to indicate a synchronization point and a non-target area.
- the point designated by is used as a synchronization point, and the state copy process is speeded up by not copying the state of the unused area, and the other configuration is the same as in the first embodiment.
- information relating to the API is created in advance in the information relating to the execution of the application 108.
- a point indicating the end of an application or a point indicating a switching point of a processing phase is set as a calling point, and an API 1101 indicating that the calling point is a synchronization point is set in advance.
- an API 1101 indicating that the calling point is a synchronization point is set in advance.
- An API 1102 indicating the presence is created in advance by the function “register_unused”.
- the system virtualization processing unit When the API 1101 and the API 1102 are created in the application 108, when the application 108 becomes the API 1101 in the process of processing, the system virtualization processing unit indicates that the calling point is a synchronization point by an API call. And instructing that the API A 1102 is a non-target area different from the copy target area.
- the system virtualization processing unit 105 determines that it is a synchronization point by calling an API, and at this synchronization point, from the storage area 400 of the memory, it is excluded from the target indicated by the API ⁇ ⁇ 1102.
- the storage area 400 of the memory includes the use area 401 of the OS 106 and the use area 403 of the second application (AP # 2). Only the stored information (for example, an application program scheduled to be used after the synchronization point) is extracted, and the extracted information is copied from the active computer 101 to the standby computer 102 as information to be copied that is necessary for continuing the processing. .
- this API call point in response to an API call from the application 108, this API call point is set as a synchronization point, and at this synchronization point, from the storage area 400 of the memory, the non-target area indicated by the API 1102 Since only the information of different copy target areas is extracted, and the extracted information is copied from the active computer 101 to the standby computer 102 as the copy target information necessary for the processing continuation, information necessary for the processing continuation is obtained. Copy processing can be speeded up, and this contributes to improvement in execution performance of a highly reliable computer system.
- the present invention is used to improve the performance required for state copying between the active computer 101 and the standby computer 102 in a highly reliable computer system composed of the active computer 101 and the standby computer 102. it can.
Abstract
Description
102・・・待機系計算機
103・・・結合網
104、114・・・ハードウェア
105、115・・・システム仮想化処理部
106、116・・・OS
107、117・・・アプリケーション仮想化処理部
108、118・・・アプリケーション
109、119・・・管理OS
110・・・状態複写処理部
Claims (10)
- 入出力装置と記憶装置および処理装置を含む運用系ハードウェア資源と、前記記憶装置に格納されて、前記運用系ハードウェア資源を動作させるための運用系ソフトウェア資源とを有する運用系計算機と、
前記運用系ハードウェア資源に相当する待機系ハードウェア資源と前記運用系ソフトウェア資源に相当する待機系ソフトウェア資源とを有し、前記運用系計算機と結合網を介して接続された待機系計算機とを備え、
前記運用系ソフトウェア資源は、複数のプログラムを有し、前記複数のプログラムに対して、前記ハードウェア資源を仮想化するとともに、前記複数のプログラムの状態を監視して、前記記憶装置の情報を処理するシステム仮想化処理部を備え、
前記システム仮想化処理部は、前記プログラムの実行状態を監視して、前記待機系計算機と状態同期を行うための同期点を検出するとともに、前記検出した同期点において前記記憶装置から処理継続に必要な複写対象の情報を抽出し、前記抽出した複写対象の情報を前記結合網を介して待機系計算機に転送する、計算機システム。 - 前記システム仮想化処理部は、前記プログラムのうち前記記憶装置に格納された複数のアプリケーションプログラムの実行状態を監視して、前記複数のアプリケーションプログラムの中のいずれかのアプリケーションプログラムの終了点を前記同期点として検出し、前記記憶装置の中から、前記複写対象の情報として、前記同期点後に使用予定のアプリケーションプログラムを抽出する、請求項1に記載の計算機システム。
- 前記システム仮想化処理部は、前記プログラムのうち前記記憶装置に格納されたアプリケーションプログラムを構成する複数の処理フェーズの実行状態を監視して、前記複数の処理フェーズの中のいずれかの処理フェーズの終了点を前記同期点として検出し、前記記憶装置の中から、前記複写対象の情報として、前記同期点後に使用予定の処理フェーズに属する情報を抽出する、請求項1に記載の計算機システム。
- 前記運用系ソフトウェア資源は、前記プログラムのうちアプリケーションプログラムを格納する記憶領域の中からガーベージコレクション対象のデータを収集し、前記記憶領域を、前記アプリケーションプログラムで使用するデータを格納する使用中データ格納領域と前記収集したガーベージコレクション対象のデータを格納するための未使用データ格納領域とに分けて構成するアプリケーション仮想化処理部を備え、
前記システム仮想化処理部は、前記アプリケーション仮想化処理部により、前記記憶領域の中に前記未使用データ格納領域が確定された時点を前記同期点とし、前記記憶装置から前記複写対象の情報として、前記使用中データ格納領域に格納されたデータを抽出する、請求項1に記載の計算機システム。 - 前記システム仮想化処理部は、前記プログラムのうち前記記憶装置に格納された複数のアプリケーションプログラムの実行状態を監視して、前記各アプリケーションプログラムに記載されたアプリケーション・プログラマブル・インタフェースで指示された同期点を前記同期点として検出し、前記記憶装置から前記複写対象の情報として、アプリケーション・プログラマブル・インタフェースで指示された情報以外の情報であって、前記同期点後に使用予定のアプリケーションプログラムを抽出する、請求項1に記載の計算機システム。
- 入出力装置と記憶装置および処理装置を含む運用系ハードウェア資源と、前記記憶装置に格納されて、前記運用系ハードウェア資源を動作させるための運用系ソフトウェア資源とを有する運用系計算機と、
前記運用系ハードウェア資源に相当する待機系ハードウェア資源と前記運用系ソフトウェア資源に相当する待機系ソフトウェア資源とを有し、前記運用系計算機と結合網を介して接続された待機系計算機とを備え、
前記運用系ソフトウェア資源は、複数のプログラムを有し、前記複数のプログラムに対して、前記ハードウェア資源を仮想化するとともに、前記複数のプログラムの状態を監視して、前記記憶装置の情報を処理するシステム仮想化処理部を備え、
前記システム仮想化処理部は、前記プログラムの実行状態を監視して、前記待機系計算機と状態同期を行うための同期点を検出するステップと、
前記ステップで検出した同期点において前記記憶装置から処理継続に必要な複写対象の情報を抽出するステップと、
前記ステップで抽出した複写対象の情報を前記結合網を介して待機系計算機に転送するステップを実行する、計算機システムの構成方法。 - 前記システム仮想化処理部は、
前記プログラムのうち前記記憶装置に格納された複数のアプリケーションプログラムの実行状態を監視して、前記複数のアプリケーションプログラムの中のいずれかのアプリケーションプログラムの終了点を前記同期点として検出するステップと、
前記記憶装置の中から、前記複写対象の情報として、前記同期点後に使用予定のアプリケーションプログラムを抽出するステップを実行する、請求項6に記載の計算機システムの構成方法。 - 前記システム仮想化処理部は、
前記プログラムのうち前記記憶装置に格納されたアプリケーションプログラムを構成する複数の処理フェーズの実行状態を監視して、前記複数の処理フェーズの中のいずれかの処理フェーズの終了点を前記同期点として検出するステップと、
前記記憶装置の中から、前記複写対象の情報として、前記同期点後に使用予定の処理フェーズに属する情報を抽出するステップを実行する、請求項6に記載の計算機システムの構成方法。 - 前記運用系ソフトウェア資源は、
前記プログラムのうちアプリケーションプログラムを格納する記憶領域の中からガーベージコレクション対象のデータを収集し、前記記憶領域を、前記アプリケーションプログラムで使用するデータを格納する使用中データ格納領域と前記収集したガーベージコレクション対象のデータを格納するための未使用データ格納領域とに分けて構成するアプリケーション仮想化処理部を備え、
前記システム仮想化処理部は、
前記アプリケーション仮想化処理部により、前記記憶領域の中に前記未使用データ格納領域が確定された時点を前記同期点とするステップと
前記記憶装置から前記複写対象の情報として、前記使用中データ格納領域に格納されたデータを抽出するステップを実行する、請求項6に記載の計算機システムの構成方法。 - 前記システム仮想化処理部は、
前記プログラムのうち前記記憶装置に格納された複数のアプリケーションプログラムの実行状態を監視して、前記各アプリケーションプログラムに記載されたアプリケーション・プログラマブル・インタフェースで指示された同期点を前記同期点として検出するステップと、
前記記憶装置から前記複写対象の情報として、アプリケーション・プログラマブル・インタフェースで指示された情報以外の情報であって、前記同期点後に使用予定のアプリケーションプログラムを抽出するステップを実行する、請求項6に記載の計算機システムの構成方法。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/201,579 US20120030504A1 (en) | 2009-03-19 | 2009-11-05 | High reliability computer system and its configuration method |
CN2009801567219A CN102317921A (zh) | 2009-03-19 | 2009-11-05 | 高可靠性计算机系统及其构成方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2009067299A JP5352299B2 (ja) | 2009-03-19 | 2009-03-19 | 高信頼性計算機システムおよびその構成方法 |
JP2009-067299 | 2009-03-19 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2010106593A1 true WO2010106593A1 (ja) | 2010-09-23 |
Family
ID=42739267
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2009/005872 WO2010106593A1 (ja) | 2009-03-19 | 2009-11-05 | 高信頼性計算機システムおよびその構成方法 |
Country Status (4)
Country | Link |
---|---|
US (1) | US20120030504A1 (ja) |
JP (1) | JP5352299B2 (ja) |
CN (1) | CN102317921A (ja) |
WO (1) | WO2010106593A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103703427A (zh) * | 2011-07-20 | 2014-04-02 | 飞思卡尔半导体公司 | 同步第一处理单元和第二处理单元的处理装置和方法 |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5136593B2 (ja) * | 2010-05-21 | 2013-02-06 | 横河電機株式会社 | 分析装置 |
JP6089427B2 (ja) * | 2012-03-30 | 2017-03-08 | 日本電気株式会社 | フォールトトレラントサーバ、デフラグ方法、およびプログラム |
US20150227599A1 (en) * | 2012-11-30 | 2015-08-13 | Hitachi, Ltd. | Management device, management method, and recording medium for storing program |
WO2014165976A1 (en) * | 2013-04-10 | 2014-10-16 | Berryman Jeremy | Multitasking and screen sharing on portable computing devices |
CN110603444A (zh) * | 2017-06-08 | 2019-12-20 | 株式会社岛津制作所 | 分析系统、控制器以及数据处理装置 |
JP2020160653A (ja) * | 2019-03-26 | 2020-10-01 | 日本電気株式会社 | 情報処理システム、仮想マシン、物理マシン、情報処理方法、及びプログラム |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0736761A (ja) * | 1993-07-21 | 1995-02-07 | Nippon Telegr & Teleph Corp <Ntt> | 外部記憶装置のオンライン高信頼複写処理方法 |
JPH07306796A (ja) * | 1994-05-13 | 1995-11-21 | Fuji Electric Co Ltd | 二重化制御システムのデータ同期転写方法 |
JPH08328891A (ja) * | 1995-06-02 | 1996-12-13 | Mitsubishi Electric Corp | 待機冗長化構成の二重化システム |
JPH11259326A (ja) * | 1998-03-13 | 1999-09-24 | Ntt Communication Ware Kk | ホットスタンバイシステムおよびホットスタンバイシステムにおける自動再実行方法およびその記録媒体 |
JP2001297011A (ja) * | 2000-04-14 | 2001-10-26 | Nec Soft Ltd | 無停止ジョブ起動方法及び無停止ジョブ起動システム |
JP2002215416A (ja) * | 2001-01-19 | 2002-08-02 | Mitsubishi Electric Corp | フォールトトレラント計算機システム |
JP2003296133A (ja) * | 2002-04-05 | 2003-10-17 | Fuji Electric Co Ltd | コントローラ |
JP2005148910A (ja) * | 2003-11-12 | 2005-06-09 | Nec Saitama Ltd | データ二重化装置及び方法 |
JP2006072591A (ja) * | 2004-09-01 | 2006-03-16 | Hitachi Ltd | 仮想計算機制御方法 |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1993009494A1 (en) * | 1991-10-28 | 1993-05-13 | Digital Equipment Corporation | Fault-tolerant computer processing using a shadow virtual processor |
US6044475A (en) * | 1995-06-16 | 2000-03-28 | Lucent Technologies, Inc. | Checkpoint and restoration systems for execution control |
US6360331B2 (en) * | 1998-04-17 | 2002-03-19 | Microsoft Corporation | Method and system for transparently failing over application configuration information in a server cluster |
US6421739B1 (en) * | 1999-01-30 | 2002-07-16 | Nortel Networks Limited | Fault-tolerant java virtual machine |
US7093086B1 (en) * | 2002-03-28 | 2006-08-15 | Veritas Operating Corporation | Disaster recovery and backup using virtual machines |
WO2007002398A2 (en) * | 2005-06-24 | 2007-01-04 | Syncsort Incorporated | System and method for virtualizing backup images |
US20070094659A1 (en) * | 2005-07-18 | 2007-04-26 | Dell Products L.P. | System and method for recovering from a failure of a virtual machine |
CN101593136B (zh) * | 2008-05-30 | 2012-05-02 | 国际商业机器公司 | 使得计算机具有高可用性的方法和计算机系统 |
US8145945B2 (en) * | 2010-01-04 | 2012-03-27 | Avaya Inc. | Packet mirroring between primary and secondary virtualized software images for improved system failover performance |
-
2009
- 2009-03-19 JP JP2009067299A patent/JP5352299B2/ja not_active Expired - Fee Related
- 2009-11-05 CN CN2009801567219A patent/CN102317921A/zh active Pending
- 2009-11-05 US US13/201,579 patent/US20120030504A1/en not_active Abandoned
- 2009-11-05 WO PCT/JP2009/005872 patent/WO2010106593A1/ja active Application Filing
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0736761A (ja) * | 1993-07-21 | 1995-02-07 | Nippon Telegr & Teleph Corp <Ntt> | 外部記憶装置のオンライン高信頼複写処理方法 |
JPH07306796A (ja) * | 1994-05-13 | 1995-11-21 | Fuji Electric Co Ltd | 二重化制御システムのデータ同期転写方法 |
JPH08328891A (ja) * | 1995-06-02 | 1996-12-13 | Mitsubishi Electric Corp | 待機冗長化構成の二重化システム |
JPH11259326A (ja) * | 1998-03-13 | 1999-09-24 | Ntt Communication Ware Kk | ホットスタンバイシステムおよびホットスタンバイシステムにおける自動再実行方法およびその記録媒体 |
JP2001297011A (ja) * | 2000-04-14 | 2001-10-26 | Nec Soft Ltd | 無停止ジョブ起動方法及び無停止ジョブ起動システム |
JP2002215416A (ja) * | 2001-01-19 | 2002-08-02 | Mitsubishi Electric Corp | フォールトトレラント計算機システム |
JP2003296133A (ja) * | 2002-04-05 | 2003-10-17 | Fuji Electric Co Ltd | コントローラ |
JP2005148910A (ja) * | 2003-11-12 | 2005-06-09 | Nec Saitama Ltd | データ二重化装置及び方法 |
JP2006072591A (ja) * | 2004-09-01 | 2006-03-16 | Hitachi Ltd | 仮想計算機制御方法 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103703427A (zh) * | 2011-07-20 | 2014-04-02 | 飞思卡尔半导体公司 | 同步第一处理单元和第二处理单元的处理装置和方法 |
Also Published As
Publication number | Publication date |
---|---|
CN102317921A (zh) | 2012-01-11 |
US20120030504A1 (en) | 2012-02-02 |
JP5352299B2 (ja) | 2013-11-27 |
JP2010218481A (ja) | 2010-09-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5352299B2 (ja) | 高信頼性計算機システムおよびその構成方法 | |
US10628273B2 (en) | Node system, server apparatus, scaling control method, and program | |
USRE47852E1 (en) | Snapshot and replication of a multi-stream application on multiple hosts at near-sync frequency | |
US9262212B2 (en) | Systems and methods for migrating virtual machines | |
US9104624B2 (en) | Systems, methods, and computer program products for instant recovery of image level backups | |
EP2955637B1 (en) | Method, system and apparatus for generating snapshot | |
US7523344B2 (en) | Method and apparatus for facilitating process migration | |
US8015431B2 (en) | Cluster system and failover method for cluster system | |
US9594522B2 (en) | Backup method and information processing apparatus | |
JP5032191B2 (ja) | サーバ仮想化環境におけるクラスタシステム構成方法及びクラスタシステム | |
US9489230B1 (en) | Handling of virtual machine migration while performing clustering operations | |
WO2016121830A1 (ja) | 仮想ネットワークファンクション管理装置、システム、ヒーリング方法及びプログラム | |
JP2014503904A (ja) | 仮想計算機のクラスタを操作するための方法、装置、コンピュータ・プログラム、およびコンピュータ・プログラム製品 | |
US20070288532A1 (en) | Method of updating an executable file for a redundant system with old and new files assured | |
WO2012049794A1 (ja) | 分散処理装置及び分散処理システム | |
CN105471994B (zh) | 一种控制方法及装置 | |
CN108469996A (zh) | 一种基于自动快照的系统高可用方法 | |
JP5933356B2 (ja) | コンピュータシステム | |
WO2011116672A1 (zh) | 为共享代码段打补丁的方法及装置 | |
Ong et al. | VCCP: A transparent, coordinated checkpointing system for virtualization-based cluster computing | |
EP4002811A1 (en) | Data synchronization method and apparatus | |
KR102472878B1 (ko) | 가상 머신 환경에서의 블록 커밋 방법 및 그 방법을 수행하는 가상화 시스템 | |
US10613789B1 (en) | Analytics engine using consistent replication on distributed sites | |
JP7000797B2 (ja) | 起動管理装置、起動管理システム、起動管理方法、および、起動管理プログラム | |
JP5544516B2 (ja) | 高可用サーバシステム、高可用サーバシステムの障害時復旧方法、および高可用サーバ |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200980156721.9 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09841800 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13201579 Country of ref document: US |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 09841800 Country of ref document: EP Kind code of ref document: A1 |