WO1996030824A1 - Procede de sauvegarde de donnees, appareil a disque miroir et commande de ce dernier - Google Patents

Procede de sauvegarde de donnees, appareil a disque miroir et commande de ce dernier Download PDF

Info

Publication number
WO1996030824A1
WO1996030824A1 PCT/JP1996/000837 JP9600837W WO9630824A1 WO 1996030824 A1 WO1996030824 A1 WO 1996030824A1 JP 9600837 W JP9600837 W JP 9600837W WO 9630824 A1 WO9630824 A1 WO 9630824A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
scsi
circuit
same
storage devices
Prior art date
Application number
PCT/JP1996/000837
Other languages
English (en)
Japanese (ja)
Inventor
Atushi Fukumoto
Original Assignee
Atushi Fukumoto
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Atushi Fukumoto filed Critical Atushi Fukumoto
Priority to AU51220/96A priority Critical patent/AU5122096A/en
Publication of WO1996030824A1 publication Critical patent/WO1996030824A1/fr

Links

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
    • G06F11/2053—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 persistent mass storage functionality or persistent mass storage control functionality is redundant
    • G06F11/2056—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 persistent mass storage functionality or persistent mass storage control functionality is redundant by mirroring
    • G06F11/2071—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 persistent mass storage functionality or persistent mass storage control functionality is redundant by mirroring using a plurality of controllers

Definitions

  • HDD hard disk drive
  • the two HDDs used for disk mirroring in 1), including the disk channel, are separate, and the same data is written into each of them and the data is duplicated to increase the reliability. Things. In this case, there is also a dedicated duplexing mirroring device.
  • the performance required of the original backup method must be a method that can secure at least one data regardless of the situation of the computer system (hardware) where data power was used. . Therefore, the backup method according to the present invention has been devised to reliably back up data even in the event of a disaster as described above.
  • a backup method in a computer network system when data is written and stored in the system, the same data is always distributed and stored in a plurality of storage devices. It is characterized by relational data distribution, not only being distributed, but also to more secure locations.
  • a control method that allows multiple SCSI storage devices to perform the same operation at the same time using a different method from the conventional mirroring device, and this method makes it possible to supply a very inexpensive mirroring device. It is characterized by. Simultaneous and distributed data storage operations directed to multiple, more secure, remote locations ensure that the probability of simultaneous loss of these distributed stored data is infinitely close to zero, ensuring Data backup power is possible.
  • the supply of inexpensive mirroring devices enables the mirroring device to be used for data at the end of the network system, thereby improving the backup performance in normal times.
  • FIG. 1 is an example of a computer system for explaining the concept of the present invention.
  • FIG. 2 is an example of a network system for explaining a specific example of the present invention.
  • FIG. 3 is an example of a display screen of the server software when softwareization is performed in order to use the backup method according to the present invention.
  • Figure 4 shows an example of the display screen when setting the data destination in the server software.
  • Figure 5 shows an example of the server software setting display screen on the S2 server.
  • Fig. 6 shows an example of the screen for setting the data gi destination of the server software on the S2 server.
  • Figure 7 shows an example of the server software setting display screen on the S3 server.
  • Fig. 8 shows an example of the data transfer destination setting display screen of the server software on the S3 server.
  • FIG. 9 is an example of a display screen of a client setting software when the backup method according to the present invention is implemented in software.
  • Figure 10-1 is a circuit diagram of an asynchronous timing control circuit.
  • Figure 10-2 shows an example of input / output signals of the asynchronous timing control circuit.
  • Figure 11 shows an example of a circuit in which multiple units are controlled by an asynchronous timing control circuit.
  • Figure 12 shows a circuit example of the REQ pulse detection circuit.
  • FIG. 13 is a block diagram for performing control for reading data in the synchronous data transfer method.
  • Figure 14 is a mask circuit timing diagram.
  • Figure 15 shows a circuit example of the level detection circuit, mask circuit, and REQU pulse generation circuit.
  • FIG. 16 is a block diagram for performing control for writing data in the synchronous data transfer method.
  • Fig. 17 shows an example of a circuit board for bidirectional input / output buses.
  • the other one is used.
  • the other one in this case is a remote storage device that is always outside the local loop, and that the harming operations performed by these two units are performed simultaneously or within a time that can be considered simultaneous. Is a point.
  • Fig. 2 is an example for explaining the client-server system that is frequently used at present.
  • Each system server of system A, system B, and system C has RA I DA, B, It is assumed that C is used, and each server can send and receive data mutually by L1 and L2. Multiple terminals as clients are connected to each server.
  • Server S1 in system A uses HDD-1 as a shared disk Al, A2, A3 and HDD-4 as a shared disk A4, A5 and HDD—A6, A7, A8, and A9 that use 3 as a shared disk.
  • Server S2 in system B uses HDD4 as a shared disk.
  • B1, B2 and HDD5 that use HDD5 as a shared disk.
  • B4, B5 and HDD6 are shared disks
  • B6, B7, B8 Force Server S3 in system C uses HDD7 as a shared disk
  • C1 and HDD8 are shared disks
  • C3, C4, C5 and HDD 9 are shared disks.
  • C6, C7 and HDD 10 are dedicated disks.
  • each of the servers 1, 2, and 13 exchanges data with a plurality of servers in other locations.
  • Data is simultaneously distributed by the ⁇ 3 ⁇ 4 path to perform data sharing, so that, for example, when the data of the server 1 becomes unavailable, the data of the server 12 via the feit path L3 or the feii path
  • server 3 is located in a remote place such as Kyushu, all data will be lost at the same time even in the event of an earthquake or other disaster: It seems unlikely.
  • This secured data has already been transferred to the server side. Only sent data. In other words, as shown in Fig.
  • S 1 passes through the route 1 and feied this data to S 2 or S 3 or both S 2 and S 3 via L 3.
  • 3 performs the same data duplication to each of the RA IDs of SI, S2, and S3 by harming the data to RA ID-C.
  • this data is written to HDD 4 and simultaneously transferred to S2 and B5.
  • B5 When B5 receives this, it writes this data to HDD 5, and S2 writes this data to RA ID-B, and at the same time, sends data to SI and S3 via feL3 and L2.
  • S 1 damages this data to RA ID-A
  • S 3 writes this data to RA ID DC.
  • the CI receives the data from the CI and receives the data.
  • the C 1 can also write the data to the HDD 7.
  • the operation of each terminal is whether or not to perform data feii at the same time as writing.
  • the software that determines the operation of each terminal is set by setting in advance where the data transfer is to be performed. It is only necessary to incorporate the method of the present invention into software because it is a multitasking method for personal computers and workstations currently used. Also, rather than embedding this software in each application software, it is the same as the application software used.
  • the software used on the server side will be described. For each setting, it is assumed that the mouse cursor is moved to the position of the setting item using the keyboard input or the mouse controller, and the click operation is performed.
  • the next setting is the setting of the relationship between the number of groups with the client and its main drive connected to the server S1, and in this example, the number of groups of the client is 3.
  • A2, A3, HDD lo Group B at the A4,
  • A5, HDD 2 0 Group C place to A6, A7, A8, and Enter A9, HDD3.
  • the software for exchanging the setting data and creating a map of the network system is activated by moving the mouse cursor to the location of the mouth map shown in Fig. 3 and clicking.
  • each server to know where and what drives are located by having a map of the network system.
  • the next setting after creating this map is to set whether or not to write the same data to other remote disks when data is written to the local drive of the server, and If this is set to NO, the server will not retransmit data, and if set to YES, all disk names (storage device names) used on the network system will be displayed on the screen as shown in Fig. 4. Is displayed. Move the mouse cursor to the position of the displayed disk name and click the mouse to set the drive to which the data is to be transferred and to perform the damage. This setting can be performed for multiple units at the same time, and by doing this, the same data can be simultaneously written to multiple locations and written.
  • FIGs 5, 6, 7, and 8 show examples of setting software for setting the other servers S2 and S3 used in Figure 2.
  • the setting of the software used on the client side shall be performed after the setting of the software on the server side and the completion of installation.
  • the map data on the server side is automatically imported.
  • the setting software on the client side can also grasp all the drive names used on the network system.
  • the setting screen shown in Fig. 9 is displayed. This is an example of a screen for setting the client software. In this case as well, each setting is performed by keyboard input or using the mouse controller to move the mouse cursor to the position of the setting item and click. It is assumed.
  • the mirroring method is a very effective method of backing up data. It is.
  • the software includes the functions of disk mirroring and disk duplexing, the functions are effective.
  • the storage device on the server side is used, or the software itself does not have a mirroring function, so the software itself is often limited.
  • the mirroring device itself is used. Since it is very expensive, it is almost impossible to use it as a storage device at the end of a computer system.
  • the present inventors have devised a control method that enables the same data to be simultaneously written or read to a plurality of SCS I storage devices according to the present invention, and a control method using the same.
  • a mirroring device By using the method according to the invention, a mirroring device can be supplied at a very low price.
  • this method will be described.
  • the SCSI system when an HDD or an optical disk drive is connected to a computer device as an external device, the most frequently used interface is the SCSI system.
  • the SCSI system When inputting / outputting data in this SCSI system, there are a synchronous transfer system and an asynchronous fe3 ⁇ 4 system, and almost all currently sold SCSI devices are compatible with the synchronous system capable of high-speed data transfer. is there.
  • the synchronous transfer method exchanges synchronous messages between the initiator and the target when the computer is turned on or resets, and confirms the speed and offset values. Is what is done.
  • a major feature of the SCSI device used in the SCSI system is that it is an intelligent device. This is because the control of the head position of the disk drive, which was performed by the CPU of the computer in the conventional floppy disk, does not need to be performed at all, and only the command is transferred. As a result, the burden on the CPU on the computer side is reduced. This is a very efficient method when HDDs are used in a normal way. However, in order to do this, each SCSI device has to use a CPU. Will be built-in during command execution. The operation is performed at an independent timing.
  • the HDD processes commands from the computer at its own timing and controls the SCSI bus.
  • the HDD is used as the storage device of SCS I, and the method of writing and reading by making the minimum number of two perform the same operation at the same time is explained.
  • FIG. 10-1 is an example of a circuit when two devices are used
  • Figure 10-2 is an example of the input / output signals of this circuit.
  • This circuit is made up of 5 circuits, and the input signals A and B are input with the output signals REQ, BUSY, MSG, C / D and I / O of HDD-A and HDD-B. If connected to the SCSI bus on the initiator side, the SCS I bus will be driven only after the two inputs A and B or the rain switch are simultaneously turned on in each circuit, that is, the two HDDs A and B have the same HDD.
  • HDD-A After entering the operating state, that state is output to other SCS I devices, mainly the initiator.
  • HDD-A outputs a REQ signal for some command and HDD-B outputs a REQ signal with a delay
  • HDD-B outputs a REQ signal for the initiator-side SCSI bus REQ bus.
  • the initiator returns an ACK pulse as a response signal because it is not driven until the two HDDs output the REQ signal.Therefore, only one of the HDDs is activated. absolutely not.
  • the operation of the two HDDs must be in the same state, and the SCS I bus is not driven. When a response signal is received, it is performed in the same operation state.
  • the signals ATN, ACK and RST output from the initiator side in the control signal line are It is only necessary to connect to each HDD directly or through a buffer.
  • the data bus and the parity bus detect the direction of input / output by the I / O signal of the SCSI bus on the initiator side, and the data is transferred from the initiator side to the HDD. If the data is input to both HDDs, this data should be input to both HDDs, and if data is output from the HDD side, only one arbitrary data should be output to the SCS I bus on the initiator side. good.
  • the two HDDs are set not only with the same ID number, but also with the same partitioning method, area reservation capacity and physical sector length, and the same data contents. I have to put it.
  • both HDDs output the same data, it is possible to have only one of them perform the reading operation and transfer that data to the initiator. It is unknown whether or not the data is actually read. For this reason, in the present invention, even when data is read from the HDD, the same operation is simultaneously performed on the two HDDs, and only one of the data is transmitted to the initiator. . At this time, which data of the two HDDs is used depends on the data of the HDD with the delayed response signal output for the data read command.
  • the reason for this is to allow the two HDDs to perform the same operation at the same time, including the movement and position of the heads, and at the same time, by deciding which HDD data to use by this method, By transferring the actually read data to the initiator and using it by random selection,
  • the relationship between the REQ signal and the ACK signal when performing the data read operation in the synchronous data transfer method will be briefly described.
  • the HDD that receives the data read command from the initiator side will not exceed the offset value.
  • Data can be transferred to the initiator in advance of the ACK pulse in the evening synchronized with the REQ pulse, and upon receiving this normally, the initiator outputs an ACK pulse to the HDD as the next data request signal That is.
  • This number of ACK pulses never exceeds the number of REQ pulses output by the HDD, and if an ACK pulse more than the REQ pulse is returned, the HDD will The operation is stopped assuming that an error has occurred. Therefore, in the present invention, it is effective to control the R EQ pulse and the ACK pulse.
  • each of the two HDDs tries to send data of less than the offset value determined in the synchronous message exchange to the initiator together with the REQ pulse to the initiator. However, a random time shift occurs at this output timing. At this time, the REQ pulse and data output earlier of the two HDDs are not output to the SCSI bus, but the REQ pulse and data output later are selected and output to the SCSI bus. .
  • the REQ pulse detection circuit detects which of the two HDDs performed the output operation first, and outputs the REQ pulse signal output with a delay to the SCSI bus.
  • the signals SEL-A and SEL-B obtained by the REQ pulse detection circuit select the data of the HDD used and control signals other than REQ BUS Y. MSG, C / D, IZO and output them to the bus This is the select signal used to FIG. 12 shows an example of the REQ pulse detection circuit.
  • other control signals The explanation of the processing and control method of the REQ pulse output by the other HDD and the signal output by the initiator, which is not performed overnight, indicates that the HDD outputs the AC of the output REQ pulse JiLt as described above. When a K-pulse force is input, an error state occurs. Therefore, in the present invention, a control method has been devised in which two HDDs that simultaneously perform the same operation do not enter such an error state. This will be described with reference to FIGS.
  • FIG. 13 is a block diagram of a circuit for processing and controlling a REQ pulse and an ACK pulse when reading data in the synchronous data transfer method.
  • the configuration of this circuit is
  • An ACK pulse generation circuit for generating an ACK pulse to be output to the HDD, a reference clock circuit, and a mask circuit for correcting the output of the ACK pulse generation circuit.
  • a level detection circuit that detects the level of the ACK signal output from the initiator and controls the mask circuit.
  • each counter circuit When selecting and outputting the data and REQ pulse of any one of the two HDDs described above, input the REQ pulse output from the non-selected HDD to the counter circuit 1, Count this.
  • the output of the ACK pulse generation circuit which will be described later, is input to the count circuit 2 and counted.
  • the ACK pulse output from the initiator is input to the counter circuit 3 and counted.
  • the length of each counter is usually 8B IT3 ⁇ 4g. This means that the difference between the count values of the counters does not become the offset value J: except when an error occurs, and it is also detected that the count value of each power counter becomes the same when the count is over. This allows match detection to be performed without any problems.
  • Comparator circuit 1 detects the coincidence of the count values of counter circuit 1 and counter circuit 2.
  • Comparator circuit 2 is one of the counter values of counter circuit 2 and counter circuit 3! If the match detection power is detected in one or both of the comparator circuits 1 and 2, the operation of the ACK pulse generation circuit is stopped. The output of the ACK pulse generation circuit controlled by this counter circuit and comparator circuit. The number of pulses will not exceed the number of REQ pulses on either HDD. However, the ACK pulse created by the above method cannot be used directly as an ACK pulse for two HDDs.
  • the ACK pulse that the initiator outputs to the HDD is usually used not only in the number of pulses but also in the level, in order to separate the data block and inform the HDD that it is the last and prepare for the next operation. And output differently. Specifically, the ACK pulse is kept at "L" and the level is returned to the normal level after the last data is output. For this reason, even when creating an AC K pulse and inputting it to the HDD, simply using the combined number of AC K pulses will cause the two HDDs to move to the next operation at different timings. If the operation state is different, an error state will result and the operation will stop. In order to deal with this, the present invention uses the following method.
  • the two HDDs perform the same operation in which the number of read data is also controlled by the controlled ACK pulse.
  • the last data output timing of the two HDDs is also the same, and the timing of the ACK pulse generated by the ACK pulse generation circuit is also determined by the ACK pulse output by the initiator.
  • the level detection circuit detects whether or not the level of the ACK pulse output from the initiator has been output for a certain period of time, and when this is detected, the mask circuit is operated by this detection signal.
  • the mask circuit performs a mask process on the ACK pulse created by the ACK pulse generation circuit, and the waveform that follows the format of the ACK pulse of ⁇ ⁇ ⁇ ⁇ that is output by the initiator. By inputting the output of this mask circuit to two HDDs as an ACK signal, the two HDDs simultaneously shift to the next operation.
  • the timing of the operation of this mask circuit is shown in FIG. 14, and FIG. 15 shows an example of a level detection circuit and an ACK pulse generation circuit for detecting "j" for a certain time or more.
  • This method is a control method for reading when synchronous data ⁇ !
  • the ACK pulse input to the HDD, the ACK signal from the initiator, and the output signal from the HDD are not the selected signals, but the output of the timing processing circuit during asynchronous operation. Must be switched so that is output to the initiator-side bus, and the original operating state must be restored.
  • the HDD receives a damage command from the initiator
  • the HDD outputs a number of REQ pulses within the range that does not exceed the offset value.
  • the initiator outputs the same number of data and an ACK pulse synchronized with the output timing of this data to the HDD.
  • the two HDDs output REQ pulses at the same timing, there is no problem even if they are simply connected in parallel.In this case, the response of the two HDDs is the same as in the read operation.
  • an ACK pulse to be input to the HDD is created at the time of the above-described reading, and two HDDs are controlled. Is a signal that informs the number of writable data or controls and outputs a REQ pulse that is a signal requesting its transfer. Creates a REQ pulse with the number of data that can be harmed, and outputs this to the initiator's REQ bus. This controls the timing at which the initiator outputs data.
  • FIG. 16 shows a block diagram of a circuit for performing this.
  • this circuit will be described with reference to FIG. 16.
  • the circuit configuration is very similar to the control circuit for reading data described above. The difference is that there is no mask circuit and there is no level detection circuit to control it, and the signal power input to each counter circuit is different.
  • the REQ pulse generation circuit is controlled by the coincidence detection output of the comparator circuit as before, and the circuit or its operation is basically the same as before. However, it is said that it is one thing for the explanation, but it is actually " ⁇ " to use one circuit in common.
  • the counter circuit 1 and the counter circuit 3 each take the output of the £ ⁇ 3 signal of each of the two 1100s as an input signal and counts each.
  • the counter 1 circuit 2 receives an output signal of the REQ pulse generation circuit as an input, and counts the output pulse.
  • the comparator circuit A compares the count value of the counter circuit 1 with the count value of the counter circuit 2 to detect coincidence.
  • the comparator circuit compares the count value of the counter circuit 2 with the count value of the counter circuit 3 to detect coincidence.
  • the operation of the REQ pulse generation circuit shall be stopped while the match detection power ⁇ is performed by either or both of the comparator circuits A and B.
  • the number of REQ pulses output by this circuit and control is such that the smaller number of REQ pulses output from the two HDDs is output.
  • this REQ pulse generation circuit if the output of this REQ pulse generation circuit is output to the SCSI bus on the initiator side, the initiator outputs only the writable number of data and the ACK pulse for both HDDs, so the two HDDs The same data can be written at the same time without error.
  • the data output from the initiator is naturally input to both HDDs, and the output signals from the other initiators are also input to both HDDs.
  • the output signal on the HDD side only the REQ signal is input to one counter circuit as described, and the other signal output outputs the output signal of the selected HDD to the initiator bus in the data reading.
  • the output signal of the timing processing circuit during asynchronous operation is output to the s CSI bus on the initiator side.
  • this method is also a method of controlling harm during synchronous data transfer.
  • the REQ pulse input to the initiator must be switched to that of the asynchronous timing processing circuit, and the other input / output signals must be in the original operating state.
  • FIG. 17 shows a circuit example when BUSY and SEL are used as bidirectional input / output circuits.
  • the two HDDs used by the above method and circuit can be regarded as one HDD from the viewpoint of the initiator, this user can use the conventional software as it is, and the operation is completely Since it does not change, it can be used without being aware of the presence or absence of the device according to the present invention, and it is free from the troublesome work of making a knock-up copy on the FDD every time the data that has been harmed to the HDD is saved. Become. In the above description, the processing at the time of occurrence of an error is not described. Therefore, when the following description is made, two HDDs are connected to the circuit described above, and a command is given from the initiator side to both of them. Have the same data written to the HDD at the same time.
  • the same operation is performed at the same time to read data.
  • the read operation is a data state check operation because the HDD from which data is read is selected by random selection.
  • the operations of both HDDs are not the same, in which case some kind of abnormality has occurred in the HDDs. Since the operation at this time stops without transitioning to the next operation state even if either HDD fails, the user using this can immediately know that the HDD has failed. it can. Even at this time, one of the HDDs must be able to perform normal operation. In this case, the normally operating HDD is directly connected to the SC SI bus on the iniche overnight without passing through the circuit according to the present invention.
  • the user can be notified of the abnormality simply by a buzzer or lamp, etc., and one of the two HDDs can be switched with a switch to provide a circuit that can be connected to the initiator. If a parity check circuit for data is added, it is possible to detect which of the two HDDs is in an error state. Connect a normally working HDD to the initiator It is also possible to do that.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Techniques For Improving Reliability Of Storages (AREA)

Abstract

Selon le procédé de sauvegarde de données de l'invention les traitements de préservation des données s'éxécutent simultanément non seulement pour la mémoire d'une boucle locale mais également pour une mémoire d'une boucle éloignée. L'invention concerne également le procédé de commande d'un appareil à disque miroir efficace en sauvegarde des données et peu coûteux.
PCT/JP1996/000837 1995-03-31 1996-03-28 Procede de sauvegarde de donnees, appareil a disque miroir et commande de ce dernier WO1996030824A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU51220/96A AU5122096A (en) 1995-03-31 1996-03-28 Data backup method, mirror ring apparatus and its control

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP7/109899 1995-03-31
JP7109899A JPH08272666A (ja) 1995-03-31 1995-03-31 データバックアップ方法と、ミラーリング装置と、その 制御方法。

Publications (1)

Publication Number Publication Date
WO1996030824A1 true WO1996030824A1 (fr) 1996-10-03

Family

ID=14521990

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP1996/000837 WO1996030824A1 (fr) 1995-03-31 1996-03-28 Procede de sauvegarde de donnees, appareil a disque miroir et commande de ce dernier

Country Status (3)

Country Link
JP (1) JPH08272666A (fr)
AU (1) AU5122096A (fr)
WO (1) WO1996030824A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9971656B2 (en) 2010-12-13 2018-05-15 International Business Machines Corporation Instant data restoration

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6035347A (en) * 1997-12-19 2000-03-07 International Business Machines Corporation Secure store implementation on common platform storage subsystem (CPSS) by storing write data in non-volatile buffer
JP3800527B2 (ja) 2002-05-30 2006-07-26 インターナショナル・ビジネス・マシーンズ・コーポレーション ネットワークを利用したデータのバックアップ技術
JP2004302512A (ja) 2003-03-28 2004-10-28 Hitachi Ltd クラスタコンピューティングシステム、および、そのフェールオーバー方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05108276A (ja) * 1991-10-11 1993-04-30 Nec Software Kansai Ltd データ処理装置
JPH05165732A (ja) * 1991-12-13 1993-07-02 Hokkaido Nippon Denki Software Kk 二重化ハードディスク制御方式
JPH0612190A (ja) * 1992-06-25 1994-01-21 Hitachi Ltd 磁気ディスク制御方式
JPH06214853A (ja) * 1992-12-02 1994-08-05 Internatl Business Mach Corp <Ibm> データベースのバックアップおよび復元の方法およびシステム

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05108276A (ja) * 1991-10-11 1993-04-30 Nec Software Kansai Ltd データ処理装置
JPH05165732A (ja) * 1991-12-13 1993-07-02 Hokkaido Nippon Denki Software Kk 二重化ハードディスク制御方式
JPH0612190A (ja) * 1992-06-25 1994-01-21 Hitachi Ltd 磁気ディスク制御方式
JPH06214853A (ja) * 1992-12-02 1994-08-05 Internatl Business Mach Corp <Ibm> データベースのバックアップおよび復元の方法およびシステム

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9971656B2 (en) 2010-12-13 2018-05-15 International Business Machines Corporation Instant data restoration
US9983946B2 (en) 2010-12-13 2018-05-29 International Business Machines Corporation Instant data restoration

Also Published As

Publication number Publication date
JPH08272666A (ja) 1996-10-18
AU5122096A (en) 1996-10-16

Similar Documents

Publication Publication Date Title
US7793060B2 (en) System method and circuit for differential mirroring of data
US7694177B2 (en) Method and system for resynchronizing data between a primary and mirror data storage system
US6721851B2 (en) System and method for preventing sector slipping in a storage area network
US7437598B2 (en) System, method and circuit for mirroring data
US6345368B1 (en) Fault-tolerant access to storage arrays using active and quiescent storage controllers
US7464236B2 (en) Storage system and storage management method
US9916113B2 (en) System and method for mirroring data
KR100604242B1 (ko) 파일서버 기억장치
JP4387707B2 (ja) クラスタリングシステムのサイトでの双方向障害検出の為のシステム及び方法
JP3655963B2 (ja) 記憶制御装置、それを含むデータ記憶システムおよび二重ペア抑止方法
US5680580A (en) Remote copy system for setting request interconnect bit in each adapter within storage controller and initiating request connect frame in response to the setting bit
US6802023B2 (en) Redundant controller data storage system having hot insertion system and method
JP2576847B2 (ja) 記憶制御装置及び関連する方法
US8402212B2 (en) Storage system and management method of its storage medium
JP3706179B2 (ja) 二重磁気ディスク制御装置間の通信制御装置
JP4902289B2 (ja) バックアップシステム及びバックアップ方法
GB2302428A (en) Multi-media storage system
US7797571B2 (en) System, method and circuit for mirroring data
CN113342261A (zh) 伺服器与应用于伺服器的控制方法
WO1996030824A1 (fr) Procede de sauvegarde de donnees, appareil a disque miroir et commande de ce dernier
US6728818B2 (en) Dual storage adapters utilizing clustered adapters supporting fast write caches
US20080256313A1 (en) System, Method And Computer Program Product For Remote Mirroring
US7281081B1 (en) System and method for preventing sector slipping in a storage area network
US10248511B2 (en) Storage system having multiple local and remote volumes and multiple journal volumes using dummy journals for sequence control
US20120246423A1 (en) Method and System for Data Replication

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AU BR CA CN KR MX RU US

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase

Ref country code: CA