JP2009059280A - Storage control system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that it is difficult to normally execute a high bit rate application requiring real time property due to a data transfer failure in a storage device which restores data by a RAID (Redundant Arrays Independent Disks) function when a CRC error or parity error occurs. <P>SOLUTION: When a failure occurs but data are not destroyed in storage devices 1 to 5, it is detected by checking residual amount of data buffers 101 to 105 for temporarily storing data read from the storage devices 1 to 5, and an application can be continued while restoring the data by the RAID function. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a storage apparatus that includes a plurality of storage devices that accumulate various data and has improved fault tolerance through a RAID function, and more particularly to a storage control method that has improved fault tolerance during execution of a real-time application.

  As applications and data have increased rapidly, the capacity of storage devices for storing data has increased. Under such circumstances, when a failure occurs in the storage device, the loss of lost data is large, but the cost of business interruption and recovery work is also significant. Improvement of fault tolerance by RAID (Redundant Arrays Independent Disks) technology has been proposed as one of methods for avoiding system down due to such a storage device failure.

  There is one described in Patent Document 1 as a prior art that includes a plurality of storage devices and has a RAID function. When a CRC error or a parity error occurs in one of a plurality of storage devices, the storage device described in Patent Document 1 disconnects the storage device in which the error has occurred and switches the system, and the remaining storage devices The data of the storage device in which an error has occurred is restored from the parity information.

JP 2005-122338 A

  A conventional storage apparatus restores data using a RAID function when a CRC error or a parity error occurs. However, when a failure that does not lead to such an error occurs in the storage device, for example, the data has not been destroyed, but several seeks are required to read the data due to weakening of magnetic information. If a failure that requires a high-speed video occurs, the required data rate cannot be obtained, and real-time applications such as high bit-rate video applications where real-time performance is important will not be able to transfer data in time. There was a problem that normal video screening could not be continued due to falling down and decoding chip hanging.

  The present invention has been made in view of the above-described conventional problems, and can detect an occurrence of a failure that has not led to destruction of data, thereby enabling an application that requires real-time performance. An object of the present invention is to provide a storage system in which an application can be continuously executed by a RAID function.

  The storage control method according to the present invention comprises a plurality of storage devices that store data, and has a RAID function, a client that requests reading of data stored in the plurality of storage devices, and the read request of the client. A storage control unit that sequentially reads data from the plurality of storage devices and sends the data to the client, the storage device control unit monitoring a data read processing speed of each of the plurality of storage devices, and a prescribed read processing speed When a storage device that falls below is detected, the storage device is disconnected and data is sent to the client by the RAID function of another storage device.

  The storage control method according to the present invention comprises a plurality of storage devices that store data, and has a RAID function, a client that requests writing of data to be stored in the plurality of storage devices, and the write request of this client. A storage controller that sequentially writes data to the plurality of storage devices, the storage device controller monitoring the data write processing speed of each of the plurality of storage devices, and a storage that falls below the prescribed write processing speed When a device is detected, the storage device is configured to be disconnected.

  According to the present invention, the data read processing speed of each of a plurality of storage devices is monitored, and when a storage device that has fallen below the specified read processing speed is detected, the storage device is disconnected, and a RAID function by another storage device is used. Since data is transmitted to the client, there is an effect that the application can be continuously executed by the RAID function even for an application that requires real-time performance.

  In addition, according to the present invention, the data writing processing speed of each of the plurality of storage devices is monitored, and when a storage device that falls below the prescribed writing processing speed is detected, the storage device is disconnected. This has the effect of enabling high-speed data writing in applications that require high speed.

Embodiment 1 FIG.
Embodiment 1 of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a storage control system according to the first embodiment of the present invention. The upper diagram shows a state when a failure occurs, and the lower diagram shows a state when a failure is recovered. In the figure, reference numerals 1 to 5 denote storage devices that store files such as image data, and are storage devices that construct a RAID function. Reference numerals 101 to 105 denote data buffers provided corresponding to the storage devices 1 to 5, respectively, and temporarily store data read from the storage devices 1 to 5. A storage control unit 201 controls the storage devices 1 to 5 and the data buffers 101 to 105, and 301 transmits a read command to the storage control unit 201 to request reading of data stored in the storage devices 101 to 105. It is a client.

  Next, the operation of the storage control method according to the first embodiment will be described. Image data and the like are stored in the storage devices 101 to 105, and the client 301 is configured to read out the image data stored in the storage devices 101 to 105 and display the images in real time by, for example, a video device.

  That is, the client 301 issues a read command to the storage control unit 201, and the storage control unit 201 accordingly sends to the storage devices 1 to 5 storing the data specified by the read command to the storage devices 1 to 5. Then, for example, a command such as UDMA (Ultra Direct Memory Access) is issued to instruct reading of data into the data buffers 101 to 105. Data read from the corresponding storage devices 1 to 5 according to the above-described data read command is stored in the data buffers 101 to 105. The data primarily stored in the data buffers 101 to 105 are sequentially read out to the client 301 at the read response speed requested by the application, and the application is executed. The storage control unit 201 monitors the data buffers 101 to 105, and the read response speed of the storage devices 1 to 5 in response to the data read command is not lower than the read speed from the data buffers 101 to 105 to the client 301. To manage.

  Hereinafter, the recovery procedure when a failure occurs will be described with reference to FIG. Now, assume that a failure such as weakening of magnetic information occurs in the storage device 2 as shown in the upper diagram of FIG. 1, and a situation that requires several seeks to read data occurs. The storage device 2 repeats seek several times to decrease the read response speed. When the storage device 2 falls below the read speed from the data buffer 102 to the client 301, the remaining amount of data that is temporarily accumulated decreases. . The storage control unit 201 monitors the remaining amount of data in the data buffers 101 to 105 (ST11, ST12), and the storage corresponding to the data buffer 102 when the preceding data remaining amount in the data buffer 102 falls below a prescribed remaining amount. It is determined that a failure has occurred in the device 2, and the storage device 2 disconnection process is executed (ST13).

  The storage devices 1 to 5 have a RAID function, and after the storage device 2 is disconnected, the storage device 201 receives data from the storage device of the other system if the storage device has a function according to RAID1, for example. Is recovered (ST14), and the application is continuously executed as shown in the lower diagram of FIG. 1 (ST15).

  For example, if the storage device has a function conforming to RAID 5 or 6, the data stored in the storage device 2 separated from the parity information of the data recorded in the remaining storage devices 1 and 3 to 5 is restored. To recover and continue running the application.

  In addition, by combining RAID1 and RAID5 or RAID6, the system is switched by RAID1 when a failure occurs, then the disconnected storage device 2 is replaced, and recovery is performed with the recovery function by RAID5 or RAID6. Is also possible.

  As described above, in an application that requires real-time processing, the storage device can be disconnected and data can be recovered so that the processing is not affected.

Embodiment 2. FIG.
The second embodiment of the present invention will be described below with reference to the drawings. FIG. 3 is a block diagram showing the storage control method according to the second embodiment of the present invention, where the upper diagram shows the state at the time of failure occurrence and the lower diagram shows the state at the time of failure recovery. The storage control method according to the second embodiment is an embodiment in which reading from the storage devices 1 to 5 is synchronized. In other words, this embodiment employs a method in which the next data read is not performed until all necessary data read from the storage devices 1 to 5 is completed for each read unit.

  In the figure, reference numerals 1 to 5 denote storage devices that store files such as image data, and are storage devices that construct a RAID function. Reference numerals 101 to 105 denote data buffers provided corresponding to the storage devices 1 to 5, respectively, and temporarily store data read from the storage devices 1 to 5. Reference numerals 401 to 405 are provided corresponding to the storage devices 101 to 105, respectively, and a counter that counts the number of times the response processing time for the read command of the corresponding storage devices 101 to 105 exceeds a specified time. 201 is the storage device 1 to 5, a storage control unit that controls the data buffers 101 to 105 and the counters 401 to 405, 301 sends a read command to the storage control unit 201, and requests to read data stored in the storage devices 101 to 105 Client.

  Next, the operation of the storage control method according to the second embodiment will be described. That is, the client 301 issues a read command to the storage control unit 201, and the storage control unit 201 accordingly sends to the storage devices 1 to 5 storing the data specified by the read command to the storage devices 1 to 5. Then, for example, a command such as UDMA (Ultra Direct Memory Access) is issued to instruct reading of data into the data buffers 101 to 105. At this time, since reading from each storage device 1 to 5 is synchronized, reading of the next reading unit is executed when all necessary data reading from each storage device 1 to 5 is completed for each reading unit. It will be.

  Data read from the storage devices 1 to 5 according to the above-described data read command is stored in the data buffers 101 to 105. The data primarily stored in the data buffers 101 to 105 are sequentially read out to the client 301 at the read response speed requested by the application, and the application is executed. The storage control unit 201 monitors any of the data buffers 101 to 105, and the read response speed of the storage devices 1 to 5 with respect to the previous data read command is lower than the read speed from the data buffers 101 to 105 to the client 301. Manage so that nothing happens.

  The counters 401 to 405 count the number of times that the response processing time for the read command of the corresponding storage device 101 to 105 exceeds the specified time. Specifically, the counter 401 to 405 starts when the read command is issued, It has a counting means that resets when a response is returned, combines a NOT circuit at a certain count value, combines the circuit so that the output of the counting means and an AND output are output, and counts the number of times of the output It is composed.

  Hereinafter, the recovery procedure when a failure occurs will be described with reference to FIG. Now, as shown in the upper diagram of FIG. 3, it is assumed that a failure such as weakening of magnetic information occurs in the storage device 2 and a situation that requires several seeks to read data occurs. The storage device 2 repeats seek several times to decrease the read response speed. When the storage device 2 falls below the read speed from the data buffer 102 to the client 301, the remaining amount of data that is temporarily accumulated decreases. . Further, since the reading of each storage device 1 to 5 is synchronized, the other storage devices 1 and 3 to 5 wait for the reading completion of the storage device 2 and read the next reading unit, so that the data buffer 101 , 103 to 105, the remaining data amount also decreases.

  The storage control unit 201 monitors the remaining amount of data in any one of the data buffers 101 to 105 (ST21, ST22), and if any preceding data remaining amount in any data buffer falls below a specified remaining amount, It is determined that a failure has occurred in storage devices 1-5. Then, referring to the count value of each of the counters 401 to 405, the storage device having the largest count value, for example, the storage device corresponding to the counter having the largest count value in terms of time average has the most delay and has an abnormal storage. Judge as a device. Here, since a failure has occurred in the storage device 2, the count value of the counter 402 is the maximum, and it is determined that the storage device 2 corresponding thereto is abnormal, and the storage device 2 is disconnected (ST23).

  The storage devices 1 to 5 have a RAID function, and after the storage device 2 is disconnected, the storage device 201 receives data from the storage device of the other system if the storage device has a function according to RAID1, for example. Is recovered (ST24), and execution of the application is continued as shown in the lower diagram of FIG. 3 (ST25).

  For example, if the storage device has a function conforming to RAID 5 or 6, the data stored in the storage device 2 separated from the parity information of the data recorded in the remaining storage devices 1 and 3 to 5 is restored. To recover and continue running the application.

  In addition, by combining RAID1 and RAID5 or RAID6, the system is switched by RAID1 when a failure occurs, then the disconnected storage device 2 is replaced, and recovery is performed with the recovery function by RAID5 or RAID6. Is also possible.

  As described above, in an application that requires real-time processing, the storage device can be disconnected and data can be recovered so that the processing is not affected.

  In the second embodiment, one counter 401 to 405 is provided corresponding to each of the storage devices 1 to 5, but the number of times the response time for the data read command exceeds the specified time is counted. Multiple counters for each storage device 1-5, each counting the number of times response time took longer than the short prescribed response time, and counting the number of times response time took longer than the long prescribed response time The counter may be provided. In this case, first, the count values of the counters with the long response time are compared to determine the storage devices 1 to 5 in which the failure has occurred. When the count values of all the counters are the same, the short response time It is possible to determine the storage devices 1 to 5 in which a failure has occurred by comparing the count values of these counters, and it is possible to determine the storage devices 1 to 5 in which the failure has occurred with higher accuracy.

  In the second embodiment, when the remaining amount of data temporarily stored in the data buffers 101 to 105 falls below a prescribed remaining amount, a failure occurs with reference to the count values of the counters 401 to 405. However, the storage control unit 201 may refer to the count values of the counters 401 to 405 at a predetermined timing to determine the storage device in which a failure has occurred. In this case, the timing at which the storage control unit 201 refers to the count values of the counters 401 to 405 is performed when the storage control unit 201 detects any abnormality sign.

  Further, it is assumed that the storage control unit 201 constantly monitors the count values of the counters 401 to 405, and the storage corresponding to the counters 401 to 405 whose count values exceed a specified value or whose count values are the maximum for a predetermined time average. The devices 1 to 5 may be determined as storage devices in which a failure has occurred.

Embodiment 3 FIG.
Hereinafter, a third embodiment of the present invention will be described with reference to the drawings. FIG. 5 is a block diagram showing a storage control system according to the third embodiment of the present invention. The upper diagram shows a state when a failure occurs, and the lower diagram shows a state when the failure is recovered. The storage control method according to the third embodiment is a storage control method for reading data without interposing a data buffer.

  In FIG. 5, reference numerals 1 to 5 denote storage devices that store files such as image data, and are storage devices that construct a RAID function. Reference numerals 401 to 405 are provided corresponding to the storage devices 101 to 105, respectively, and a counter that counts the number of times the response processing time for the read command of the corresponding storage devices 101 to 105 exceeds a specified time. 5 and the storage control unit 301 that controls the counters 401 to 405 are clients that send a read command to the storage control unit 201 and request reading of data stored in the storage devices 101 to 105.

  Next, the operation of the storage control method according to the third embodiment will be described. That is, the client 301 issues a read command to the storage control unit 201, and the storage control unit 201 accordingly sends to the storage devices 1 to 5 storing the data specified by the read command to the storage devices 1 to 5. Then, for example, a command such as UDMA (Ultra Direct Memory Access) is issued to instruct reading of data into the data buffers 101 to 105.

  The data read from the storage devices 1 to 5 in accordance with the data read command is sequentially read to the client 301 and the application is executed.

  Hereinafter, a recovery procedure when a failure occurs will be described. Now, as shown in the upper diagram of FIG. 5, it is assumed that a failure such as weakening of magnetic information occurs in the storage device 2, and a situation that requires several seeks to read data occurs. The storage device 2 repeats seek several times, so that the read response speed decreases and the data read amount decreases.

  The storage control unit 201 calculates the read processing speed from the read processing size represented by, for example, the number of LBA (Logical Block Addressing) from the storage devices 1 to 5 and the elapsed time (may be timer processing per unit time). The storage device 2 whose processing speed falls below the specified processing speed is determined to be abnormal. Hereinafter, the procedure for specifically detecting the failure of the storage devices 1 to 5 will be described with reference to FIG. 6. The storage control unit 201 counts the number of times that the response time to the data read command exceeds the specified time. 405, among storage devices 1 to 5, the number of times that the response time with respect to the data read command exceeds the specified time is the largest, for example, the average on the time average, and the abnormal storage with the most delay Judge as a device. Here, since a failure has occurred in the storage device 2, the count value of the counter 402 is the maximum, and the corresponding storage device 2 is disconnected (ST31).

  The storage devices 1 to 5 have a RAID function, and after the storage device 2 is disconnected, the storage device 201 receives data from the storage device of the other system if the storage device has a function according to RAID1, for example. Is recovered (ST32), and the execution of the application is continued as shown in the lower diagram of FIG. 5 (ST33).

  For example, if the storage device has a function conforming to RAID 5 or 6, the data stored in the storage device 2 separated from the parity information of the data recorded in the remaining storage devices 1 and 3 to 5 is restored. To recover and continue running the application.

  In addition, by combining RAID1 and RAID5 or RAID6, the system is switched by RAID1 when a failure occurs, then the disconnected storage device 2 is replaced, and recovery is performed with the recovery function by RAID5 or RAID6. Is also possible.

  As described above, in an application that requires real-time processing, the storage device can be disconnected and data can be recovered so that the processing is not affected.

  In the third embodiment, one counter 401 to 405 is provided corresponding to each of the storage devices 1 to 5, but the number of times that the response time for the data read command exceeds the specified time is counted. Multiple counters for each storage device 1-5, each counting the number of times response time took longer than the short prescribed response time, and counting the number of times response time took longer than the long prescribed response time The counter may be provided. In this case, first, the count values of the counters with the long response time are compared to determine the storage devices 1 to 5 in which the failure has occurred. When the count values of all the counters are the same, the short response time It is possible to determine the storage devices 1 to 5 in which a failure has occurred by comparing the count values of these counters, and it is possible to determine the storage devices 1 to 5 in which the failure has occurred with higher accuracy.

Embodiment 4 FIG.
Embodiment 4 of the present invention will be described below with reference to the drawings. FIG. 7 is a block diagram showing a storage control system according to the fourth embodiment of the present invention. The upper diagram shows a state when a failure occurs, and the lower diagram shows a state when the failure is recovered. The storage control method according to the fourth embodiment is a storage control method at the time of writing to the storage device.

  In FIG. 7, reference numerals 1 to 5 denote storage devices that store files such as image data, and are storage devices that construct a RAID function. Reference numerals 401 to 405 are provided corresponding to the storage devices 101 to 105, respectively, and a counter that counts the number of times that the response processing time for the write command of the corresponding storage devices 101 to 105 exceeds a specified time. 5 and the storage control unit 301 that controls the counters 401 to 405 are clients that send a write command to the storage control unit 201 and request the storage devices 101 to 105 to write data.

  Next, the operation of the storage control method according to the fourth embodiment will be described. That is, the client 301 issues a write command to the storage control unit 201, and the storage control unit 201 in response to the storage devices 1 to 5 specified by the write command to the storage devices 1 to 5, for example, UDMA ( A command such as Ultra Direct Memory Access) is issued to instruct data writing. Since the storage devices 101 to 105 are storage apparatuses that construct a RAID function, RAID processing is performed on the plurality of storage devices 101 to 105 and distributed writing is performed.

  Hereinafter, a recovery procedure when a failure occurs will be described. Now, as shown in the upper diagram of FIG. 7, it is assumed that a failure such as weakening of magnetic information occurs in the storage device 2, and a situation that requires several seeks to write data occurs. The storage device 2 repeats seek several times to reduce the write response speed.

  The storage control unit 201 calculates the write processing speed from the write processing size represented by, for example, the number of LBA (Logical Block Addressing) from the storage devices 1 to 5 and the elapsed time (may be timer processing per unit time). The storage device 2 whose processing speed falls below the specified processing speed is determined to be abnormal. That is, the storage control unit 201 refers to the counters 401 to 405 in which the number of times the response time for the data write command exceeds the specified time is counted, and among the storage devices 1 to 5, the response time for the data write command is specified. For example, it is determined that the number of times exceeding the set time is the abnormal storage device having the largest delay with the largest time average. Here, since a failure has occurred in the storage device 2, the count value of the counter 402 is the maximum, and the corresponding storage device 2 is disconnected.

  The storage devices 1 to 5 have a RAID function, and the storage control unit 201 writes to the other storage device if the storage device has a function conforming to RAID 1 after the storage device 2 is detached. Continue. For example, if the storage device has a function conforming to RAID 5 or 6, parity calculation is performed on the remaining storage devices 1 and 3 to 5 and writing is continued without writing to the separated storage device 2. .

  In addition, by combining RAID1 and RAID5 or RAID6, the system is switched by RAID1 when a failure occurs, then the disconnected storage device 2 is replaced, and recovery is performed with the recovery function by RAID5 or RAID6. Is also possible.

  As described above, in an application that requires real-time processing, the storage device can be disconnected and data can be recovered so that the processing is not affected.

  In the fourth embodiment, one counter 401 to 405 is provided corresponding to each storage device 1 to 5, but the number of times that the response time to the data write command exceeds the specified time is counted. Multiple counters for each storage device 1-5, each counting the number of times response time took longer than the short prescribed response time, and counting the number of times response time took longer than the long prescribed response time The counter may be provided. In this case, first, the count values of the counters with the long response time are compared to determine the storage devices 1 to 5 in which the failure has occurred. When the count values of all the counters are the same, the short response time It is possible to determine the storage devices 1 to 5 in which a failure has occurred by comparing the count values of these counters, and it is possible to determine the storage devices 1 to 5 in which the failure has occurred with higher accuracy.

  Furthermore, in each of the above-described embodiments, a configuration in which five storage devices 1 to 5 are provided is shown, but the present invention is not limited to this.

  The present invention is a storage apparatus having a RAID function, and is particularly applied to a system that executes an application that requires real-time performance.

It is the block diagram which showed the storage control system which concerns on Embodiment 1 of this invention. 3 is a flowchart showing a procedure for detecting a failure of a storage device according to the first embodiment of the present invention. It is the block diagram which showed the storage control system which concerns on Embodiment 2 of this invention. It is the flowchart which showed the procedure which detects the failure of the storage device which concerns on Embodiment 2 of this invention. It is the block diagram which showed the storage control system which concerns on Embodiment 3 of this invention. It is the flowchart which showed the procedure which detects the failure of the storage device which concerns on Embodiment 3 of this invention. It is the block diagram which showed the storage control system which concerns on Embodiment 4 of this invention.

Explanation of symbols

  1 to 5 Storage device 101 to 105 Data buffer 201 Storage control unit 301 Client 401 to 405 Counter

Claims (9)

A storage apparatus having a plurality of storage devices for storing data, having a RAID function, a client requesting to read data stored in the plurality of storage devices, and sequentially starting from the plurality of storage devices based on the read request of the client A storage control unit that reads data and sends it to the client; the storage device control unit monitors the data read processing speed of each of the plurality of storage devices, and detects a storage device that falls below the specified read processing speed; In this case, the storage control method is characterized in that the storage device is disconnected and data is sent to the client by a RAID function by another storage device. A plurality of storage devices for storing data and having a RAID function; a plurality of data buffers provided corresponding to each of the plurality of storage devices; and a plurality of data buffers for primarily storing data read from each storage device; A client that requests reading of data stored in a plurality of storage devices, and storage control that sequentially reads data from the plurality of storage devices based on the read request of the client and sends the data to the client via the plurality of data buffers The storage control unit reads each storage device so that the processing speed of reading from the plurality of storage devices to the plurality of data buffers does not fall below the speed of reading from the plurality of data buffers to the client. On the other hand, while monitoring the remaining amount of data that is primarily accumulated in the plurality of data buffers, if a data buffer that falls below the specified remaining amount is detected, the storage device corresponding to the data buffer is disconnected, A storage control method, wherein data is sent to the client by a RAID function by another storage device. A storage apparatus that includes a plurality of storage devices that store data and has a RAID function, a data buffer that primarily stores data read from the plurality of storage devices, and a request to read data stored by the plurality of storage devices Based on the read request of the client, the client, the storage control unit that reads the synchronized data from the plurality of storage devices and sends the data to the client via the data buffer, and the storage control unit Provided with a plurality of counters for counting the number of times the processing response time for the read command of each storage device exceeds a specified time, and the storage control unit receives the data buffer from the plurality of storage devices. While managing the reading of each storage device so that the reading processing speed does not fall below the reading speed from the data buffer to the client, the remaining amount of data temporarily accumulated in the data buffer is monitored and falls below the specified remaining amount. In this case, the storage control method is characterized in that the storage device corresponding to the counter having the maximum count value is disconnected from the plurality of counters, and the data is sent to the client by the RAID function of another storage device. A plurality of storage devices for storing data and having a RAID function; a plurality of data buffers provided corresponding to each of the plurality of storage devices; and a plurality of data buffers for primarily storing data read from each storage device; A client requesting to read data stored in a plurality of storage devices, and reading data synchronized with the plurality of storage devices based on the read request from the client and sending the data to the client via the plurality of data buffers A storage control unit that is provided corresponding to each of the plurality of storage devices, and includes a plurality of counters that count the number of times that the processing response time to the read command of each storage device exceeds a specified time. The unit manages the reading of each storage device so that the processing speed of reading from the plurality of storage devices to the plurality of data buffers does not fall below the speed of reading from the plurality of data buffers to the client. When the accumulated data remaining amount is monitored and falls below a prescribed remaining amount, the storage device corresponding to the counter with the maximum count value is disconnected from the plurality of counters, and the RAID function by another storage device is used to A storage control method characterized by sending data to a client. A storage apparatus having a plurality of storage devices for storing data, having a RAID function, a client requesting to read data stored in the plurality of storage devices, and sequentially starting from the plurality of storage devices based on the read request of the client A storage control unit that reads data and sends it to the client, and a plurality of counters that are provided for each of the plurality of storage devices and count the number of times that the processing response time to the read command of each storage device exceeds a specified time The storage control unit confirms the count values of the plurality of counters at a predetermined timing, disconnects the storage device corresponding to the counter with the maximum count value, and performs the RAID function by other storage devices. Storage control system, characterized by sending data to the client I. A storage apparatus having a plurality of storage devices for storing data, having a RAID function, a client requesting to read data stored in the plurality of storage devices, and sequentially starting from the plurality of storage devices based on the read request of the client A storage control unit that reads data and sends it to the client, and a plurality of counters that are provided for each of the plurality of storage devices and count the number of times that the processing response time to the read command of each storage device exceeds a specified time The storage control unit monitors the count values of the plurality of counters, disconnects the storage device corresponding to the counter whose count value has reached the specified count value, and performs RAID by another storage device. Storage control system, characterized by sending data to the client by the capacity. A storage apparatus comprising a plurality of storage devices for storing data, and having a RAID function, a client that requests writing of data to be stored in the plurality of storage devices, and in order to the plurality of storage devices based on the write request of the client A storage control unit for writing data, and the storage device control unit monitors the data write processing speed of each of the plurality of storage devices, and if a storage device that falls below the prescribed write processing speed is detected, A storage control method characterized by separation. A storage apparatus comprising a plurality of storage devices for storing data, and having a RAID function, a client that requests writing of data to be stored in the plurality of storage devices, and in order to the plurality of storage devices based on the write request of the client A storage control unit for writing data, and a plurality of counters provided corresponding to each of the plurality of storage devices, each of which counts the number of times that the processing response time to the write command of each storage device exceeds a specified time, The storage control method is characterized in that the unit confirms the count values of the plurality of counters at a predetermined timing and disconnects the storage device corresponding to the counter having the maximum count value. A storage apparatus comprising a plurality of storage devices for storing data, and having a RAID function, a client that requests writing of data to be stored in the plurality of storage devices, and in order to the plurality of storage devices based on the write request of the client A storage control unit for writing data, and a plurality of counters provided corresponding to each of the plurality of storage devices, each of which counts the number of times that the processing response time to the write command of each storage device exceeds a specified time, The storage control system characterized in that the unit monitors the count values of the plurality of counters and disconnects the storage device corresponding to the counter whose count value has reached a specified count value.

Images