JP2011192303A - Control device and control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique that improves a failure rate of a memory having a mirroring structure. <P>SOLUTION: A control device for controlling a plurality of memories having driving parts includes: a reader/writer which, on reception of a write instruction to write data to the memories, transmits a driving signal for driving the driving parts to all the memories to write the data into the memories and, on reception of a read instruction to read out the written data, transmits the driving signal to only one of the memories to read out the data from the memory; and a stop means which transmits a stop signal for stopping the driving parts to all the memories when a prescribed time passes after the end of data write or data read by the reader/writer. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a technique for controlling a drive unit of a storage device.

  An HDD (Hard Disk Drive) may fail due to wear of an internal magnetic head or a magnetic material, and the failure occurrence rate is greatly influenced by the operating time of the HDD motor. In a business terminal that operates for 24 hours such as POS, the motor operation time becomes long, and the failure occurrence rate of the HDD also increases. Therefore, a terminal that is driven for a long time is mirrored as in the configurations disclosed in Patent Documents 1 to 4 in order to maintain data and operate the terminal continuously even if one HDD fails. Often used. Mirroring is a mechanism that improves reliability of data integrity by regarding two or more HDDs as one HDD and writing the same data to all HDDs.

JP 2005-71100 A JP-A-2005-250644 JP 2006-252451 A JP 2007-115270 A

  However, although mirroring improves the reliability of data maintenance, the motor operating rate per HDD is the same as when mirroring is not used. This is because all HDDs are operated both when data is written and when data is read.

  Since the motor operating time of each HDD is the same, the failure rate per HDD is almost the same as when mirroring is not used, and the failure rate does not improve.

  An object of the present invention is to provide a technique for improving the failure rate of a mirrored storage device.

  In order to achieve the above object, the control device of the present invention is a control device that controls a plurality of storage devices having a drive unit, and when a write command instructing writing of data to the storage device is received. A drive signal for driving the drive unit to all the storage devices, writing the data to all the storage devices, and receiving a read command instructing to read the written data, a plurality of the storage devices Reading means for transmitting the drive signal to only one of the storage devices and reading the data from the storage device, and after the writing or reading of the data by the reading means is completed Stop means for transmitting a stop signal for stopping the driving unit to all the storage devices when a predetermined time has elapsed, and the reading means reads the written data. When a read instruction is received, diagnostic information indicating a result of diagnosing the state of the storage device is acquired, and any one storage device is selected based on the acquired diagnostic information, and the diagnostic information Includes energization time information indicating the energization time of the storage device, and the reading means selects the storage device having the minimum energization time indicated by the energization time information from the plurality of storage devices.

  The control method of the present invention is a control method for controlling a plurality of storage devices having a drive unit, and when a write command instructing writing of data to the storage device is received, all the storage devices A drive signal for driving the drive unit to write the data to all the storage devices, and when receiving a read command for instructing to read the written data, one of the plurality of storage devices Sending the drive signal to only one storage device, reading the data from the storage device, and stopping the drive unit when a predetermined time has elapsed after the writing or reading of the data is completed When a signal is transmitted to all of the storage devices and a read command instructing reading of the written data is received in the reading of the data, the state of the storage device As a result of diagnosis, diagnostic information including energization time information indicating the energization time of the storage device is acquired, and the energization time indicated by the energization time information from the plurality of storage devices based on the acquired diagnosis information Selects the smallest storage device.

  According to the present invention, when writing data, all the storage devices are operated, and when reading data, the drive unit of only one of the storage devices is driven. Then, when a predetermined time elapses after the reading and writing of data is completed, the drive units of all the storage devices are stopped. As a result, since only one storage device operates while data is being read, the operation time per storage device is reduced, and the failure rate of the storage device constructed by mirroring is improved.

1 is an overall view showing a configuration of a mirroring system of a first embodiment. 1 is an external view of a mirroring system according to a first embodiment. It is a flowchart which shows operation | movement of the controller of 1st Embodiment. It is a flowchart which shows the operation | movement of the reading process of 1st Embodiment. It is a flowchart which shows the write-in process of 1st Embodiment. It is a flowchart which shows the read-out process of 1st Embodiment. It is a flowchart which shows the idling process of 1st Embodiment. It is a figure which shows the operation result of the mirroring system of 1st Embodiment. It is a figure which shows the operation result of the mirroring system of 1st Embodiment. It is a figure which shows the operation result of the mirroring system of 1st Embodiment. It is a general view which shows the structure of the mirroring system of 2nd Embodiment. It is a figure which shows the structure of the diagnostic information of 2nd Embodiment. It is a flowchart which shows the read-out process of 2nd Embodiment.

(First embodiment)
A first embodiment for carrying out the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of a mirroring system 1 of the present embodiment. The mirroring system 1 is a system that improves the reliability of data integrity by writing the same data to two or more HDDs. Referring to FIG. 1, the mirroring system 1 includes an information processing apparatus 10 and a RAID (Redundant Arrays of Inexpensive Disks) unit 20.

  The information processing apparatus 10 is a computer such as a personal computer. When storing data in the RAID unit 20, the information processing apparatus 10 transmits a “Write” command to the RAID unit 20 and transmits data to be stored. In addition, when the information processing apparatus 10 reads data stored in the RAID unit 20, the information processing apparatus 10 transmits a “Read” command to the RAID unit 20. The information processing apparatus 10 transmits information indicating the address of data to be read to the RAID unit 20 and receives the read data from the RAID unit 20.

  The RAID unit 20 is an external device that provides the information processing apparatus 10 with a RAID 1 mirroring function, and includes a controller 21, an HDD 22, and an HDD 23.

  FIG. 2 is an external view of the RAID unit 20. Referring to the figure, the RAID unit 20 is connected to the information processing apparatus 10 by a USB (Universal Serial Bus) cable, a SATA (Serial Advanced Technology Attachment) cable, or the like. The RAID unit 20 has a case for housing the HDD 22 and the HDD 23. The HDD 22 and the HDD 23 are connected to the controller 21 by a SATA cable, an IDE (Integrated Drive Electronics) cable, or the like, and are stored in the RAID unit 20.

  Returning to FIG. 1, the controller 21 controls the entire RAID unit 20. The controller 21 includes a reading processing unit 211 and an idling processing unit 213.

  When the reading processing unit 211 receives a “Write” command from the information processing apparatus 10, the reading processing unit 211 transmits a standby command to the HDD 22 and the HDD 23 to operate the HDD 22 and the HDD 23. The standby command is a command for driving the drive parts of the HDD 22 and the HDD 23. The HDD 22 and the HDD 23 that have received the standby command are in a state (standby) in which the drive portion is driven to write data or read data. The reading processing unit 211 receives data to be stored from the information processing apparatus 10 and writes the received data to the HDD 22 and the HDD 23.

  Further, when receiving the “Read” command from the information processing apparatus 10, the reading processing unit 211 transmits a standby command to only one of the HDD 22 and the HDD 23 to operate. The reading processing unit 211 receives information indicating the data address from the information processing apparatus 10 and reads the data from the activated HDD. The reading processing unit 211 transmits the read data to the information processing apparatus 10. The HDD that is activated when the “Read” command is received is selected by the controller 21 in a predetermined order. For example, the HDD 22 and the HDD 23 are operated alternately.

  The idling processing unit 213 transmits a stop command to the HDD 22 and the HDD 23 to stop when a predetermined time has elapsed after the writing or reading of data by the reading processing unit 211 is completed.

  The controller 21 may be a RAID card connected to the motherboard of the information processing apparatus 10 via a PCI (Peripheral Components Interconnect) bus. The controller 21 may be a component mounted on the motherboard of the information processing apparatus 10. The information processing apparatus 10 can also realize the function of the controller 21 with software.

  In addition, the controller 21 is not limited to two, and can control three or more storage devices (such as HDDs) to provide a mirroring function.

  The HDD 22 and the HDD 23 are storage devices that store data in a platter by controlling a magnetic head or the like under the control of the controller 21 and read out the stored data. The HDD 22 and the HDD 23 have a drive unit 221 and a drive unit 231, respectively. The drive unit 221 and the drive unit 231 are driven when data is stored in the storage device 22 and the storage device 23 or when data is read out. The drive unit 221 and the drive unit 231 are, for example, a motor that rotates a platter and a motor that controls a magnetic head.

  The HDD 22 and the HDD 23 may use a storage device other than the HDD as long as the storage device has a drive part. The HDD 22 and the HDD 23 may be, for example, a hybrid HDD or a DDS (Digital Data Storage).

  Next, the operation of the controller 21 will be described with reference to FIGS. FIG. 3 is a flowchart showing a control process executed by the controller 21. This control process starts when the controller 21 is powered on or when a predetermined application is executed. Referring to the figure, the reading processing unit 211 executes the reading process (step S1), and the idling processing unit 213 executes the idling process (step S3).

  FIG. 4 is a flowchart showing the reading process. Referring to the figure, the reading processing unit 211 executes a writing process (step S10) and executes a reading process (step S20).

  FIG. 5 is a flowchart showing the writing process. Referring to the figure, the reading processing unit 211 determines whether or not a “Write” command is received from the information processing apparatus 10 (step S101).

  If the “Write” command is received (step S101: YES), the reading processing unit 211 transmits a standby command to the HDD 22 and the HDD 23. If the HDD is already in operation, the reading processing unit 211 does not transmit a standby command to the HDD. The HDD 22 and the HDD 23 receive the standby command and drive the drive unit 221 and the drive unit 231 (step S103). The reading processing unit 211 receives the data to be stored from the information processing unit 10, and writes the received data to the HDD 22 and the HDD 23 (step S105). After step S105, the reading processing unit 211 ends the writing process.

  FIG. 6 is a flowchart showing the reading process. Referring to the drawing, the reading processing unit 211 determines whether or not a “Read” command is received from the information processing apparatus 10 (step S201).

  If the “Read” command has been received (step S201: YES), the reading processing unit 211 determines whether or not data has been read from the HDD 22 when the “Read” command was received last time (step S203).

  If data was read from the HDD 22 last time (step S203: YES), the reading processing unit 211 transmits a standby command to the HDD 23 to operate the HDD 23 (step S205). If no data has been read from the HDD 22 last time (step S203: NO), the reading processing unit 211 transmits a standby command to the HDD 22 to operate the HDD 23 (step S207). If the HDD is already operating in step S205 or step S207, the reading processing unit 211 does not transmit a standby command to the HDD.

  After step S205 or step S207, the reading processing unit 211 receives information indicating the data address from the information processing apparatus 10, and reads the data from the HDD (22 or 23) in the standby state. The reading processing unit 211 transmits the read data to the information processing apparatus 10 (step S209). After step S209, the reading processing unit 211 ends the reading process.

  FIG. 7 is a flowchart showing the idling process. Referring to the figure, the idling processing unit 213 determines whether or not a data writing process (step S10) is being executed (step S31). If the data writing process is not being executed (step S31: NO), the idling processing unit 213 determines whether the data reading process (step S20) is being executed (step S33). If the data writing process is not being executed (step S33: NO), when a predetermined time has elapsed, the idling processing unit 213 transmits a stop command to the HDD 22 and the HDD 23 to stop these HDDs (step S35). . In step S35, when the HDD has already been stopped, the idling processing unit 213 does not transmit a stop command to the HDD. When the data writing process is being executed (step S31: YES), when the data reading process is being executed (step S33: YES), or after step S35, the idling processing unit 213 performs the idling process. finish.

  As described above, when data is read, only one HDD is operated, so that the operation time per HDD is reduced as compared with the case where both are operated and data is read, and as a result, the failure rate is reduced.

  Next, an example of the operation result of the mirroring system 1 will be described with reference to FIGS. When writing data to the HDD 22 and the HDD 23, the controller 21 sends a standby command to the HDD 22 and the HDD 23 to operate as shown in FIG. 8, and writes the same data to these HDDs (step S105). In the figure, arrows indicate the flow of data, and the shaded HDDs (22 and 23) are operating HDDs.

  On the other hand, when reading data from the HDD 22 or the HDD 23, the controller 21 operates only one of the HDD 22 and the HDD 23 as shown in FIG. 9 (step S205 or S207). In the figure, arrows indicate the flow of data. A shaded HDD (22) is an active HDD, and an unshaded HDD (23) is a stopped HDD.

  During idling, the controller 21 stops all HDDs (22 and 23) as shown in FIG. 10 (step S211). In the figure, HDDs without shading (22 and 23) are stopped HDDs.

  As described above, according to this embodiment, when writing data, the HDD 22 and the HDD 23 are operated, and when reading data, only one of the HDD 22 and the HDD 23 is operated. Then, after the reading / writing of data is completed, all HDDs are stopped when a predetermined time has elapsed. As a result, since only one HDD operates while data is being read, the operation time per HDD is reduced, and the failure rate of the mirrored HDD is improved.

  In addition, since the HDDs to be operated are selected according to a predetermined order, the operation times of the plurality of HDDs (22 and 23) are equalized, and the risk of failure can be distributed.

(Second Embodiment)
A second embodiment of the present invention will be described with reference to FIGS. FIG. 11 is an overall view showing the configuration of the mirroring system 2 of the present embodiment. Referring to the figure, the mirroring system 2 has the same configuration as that of the mirroring system 1 of the first embodiment, except that the HDD 22 and the HDD 23 store diagnostic information 223 and diagnostic information 233, respectively.

  The diagnosis information 223 is information in which the diagnosis result of the state of the HDD 22 for a predetermined item is digitized. The diagnostic information 223 is, for example, the S.D. M.M. A. Information acquired by the RT (Self Monitoring Analysis reporting Technology) function. FIG. 10 is a diagram showing the configuration of the diagnostic information 223. As shown in FIG. Referring to the figure, diagnostic information 223 includes information indicating item ID, item name, current value, and threshold. The item ID is a number uniquely assigned to the item name, and the item name is a diagnostic item regarding the state of the HDD 22. The diagnosis items are items that are diagnosed for the purpose of early detection of HDD failures and prediction of failures. For example, the read error rate indicating the rate of errors that occur when the HDD 22 reads data, and the overall throughput of the HDD 22 HDD processing capability indicating the power on time, POH (Power On Hours) indicating the energization time of the HDD 22, and the like. The current value is a value obtained by quantifying the real-time diagnosis result for the item name. The threshold value is a threshold value for detecting a failure or failure of the HDD 22. The diagnosis information 233 has the same configuration as the diagnosis information 223.

  Returning to FIG. 11, the controller 21 acquires diagnostic information 223 and 233 from the HDD 22 and the HDD 23. Based on the diagnosis information 223 and 233, the failure rate and failure rate of each HDD are analyzed and compared. Then, when reading data from the HDD 22 or the HDD 23, the controller 21 preferentially operates an HDD with a small failure or failure occurrence rate. For example, the controller 21 selects and operates the HDD with the least POH.

  The operation of the mirroring system 2 of this embodiment will be described with reference to FIG. This figure is a flowchart showing the reading process of the present embodiment. Referring to the figure, the reading process of the present embodiment is the same as the reading process of the first embodiment, except that step S203a is executed instead of step S203.

  If the “Read” command is received (step S201: YES), the reading processing unit 211 acquires the diagnostic information 223 and the diagnostic information 233, and compares the POH indicated by these diagnostic information. Then, the reading processing unit 211 determines whether or not the energization time (POH) of the HDD 22 is longer than the energization time of the HDD 23 (step S203a). If the energization time of the HDD 22 is longer (step S203a: YES), the reading processing unit 211 operates the HDD 23 (step S205). If the energization time of the HDD 22 is shorter (step S203a: NO), the reading processing unit 211 operates the HDD 22 (step S207).

  The controller 21 predicts the probability of occurrence of a failure or failure of each HDD based on the diagnosis result of an item other than information indicating POH or the diagnosis result of a plurality of items including POH, and the HDD having few failures and failures. It can also be configured to operate with priority.

  According to the present embodiment, since the HDD to be operated is selected based on the diagnosis information (223 and 233), it is possible to preferentially select an HDD with a low failure occurrence probability and distribute the failure risk.

  Further, since the HDD (22 or 23) having a short operation time is selected and operated, even if the energization time of each HDD is greatly different as a result of replacing the failed HDD, the operation time of the HDD is equal. The risk of failure can be distributed.

  Note that all or part of the processing shown in FIGS. 3 to 5 and FIG. 11 can also be realized by a computer executing a software program.

1, 2 Mirroring system 10 Information processing device 20 RAID unit 21 Controller 22, 23 HDD
211 Reading processing unit 213 Idling processing unit 221, 231 Driving unit 223, 233 Diagnostic information S10 to S20, S101 to S105, S201 to S211, S203a Step

Claims (4)

A control device for controlling a plurality of storage devices having a drive unit,
When a write command instructing writing of data to the storage device is received, a drive signal for driving the drive unit is transmitted to all the storage devices, and the data is written and written to all the storage devices. In addition, when a read command instructing reading of the data is received, reading means for transmitting the drive signal to only one of the plurality of storage devices and reading the data from the storage device;
Stop means for transmitting a stop signal for stopping the drive unit to all the storage devices when a predetermined time has elapsed after the writing of the data by the reading means or reading of the data is completed;
Have
When the reading means receives a read command for instructing reading of the written data, the reading means acquires diagnostic information indicating a result of diagnosing the state of the storage device, and based on the acquired diagnostic information, Select one storage device,
The diagnostic information includes energization time information indicating the energization time of the storage device, and the reading means selects the storage device with the minimum energization time indicated by the energization time information from among the plurality of storage devices. Control device.   The control device according to claim 1, wherein the storage device is an HDD (Hard Disk Drive).   The control device according to claim 1, wherein the control device is a control device corresponding to a standard of RAID (Redundant Arrays of Inexpensive Disks) 1. A control method for controlling a plurality of storage devices having a drive unit,
When receiving a write command for instructing writing of data to the storage device, a drive signal for driving the drive unit to all the storage devices is transmitted to write the data to all the storage devices,
When receiving a read command instructing reading of the written data, the drive signal is transmitted to only one of the plurality of storage devices to read the data from the storage device,
When a predetermined time elapses after the data writing or data reading ends, a stop signal for stopping the driving unit is transmitted to all the storage devices,
In the reading of the data, when a read command instructing reading of the written data is received, diagnosis information including energization time information indicating the energization time of the storage device is obtained as a result of diagnosing the state of the storage device. A control method for acquiring a storage device having a minimum energization time indicated by the energization time information from a plurality of the storage devices based on the acquired diagnostic information.

Images