KR100483412B1 - Storage system - Google Patents

Abstract

SUMMARY OF THE INVENTION An object of the present invention is to provide a control method for cooperatively processing a request for access from a higher-level device between a plurality of storage devices forming the storage system in a RAID1 storage system including a plurality of storage devices. It is done.

In order to achieve this object, the following configuration is taken. Means for each individual storage device constituting the RAID1 storage system to share an access request from a higher level device, means for determining whether or not to process the shared access request itself, and further storing from a higher level device. Means for transmitting at least the write data transmitted to the device system to the storage device to be processed.

Description

Storage system {STORAGE SYSTEM}

The present invention relates to a storage system for holding data. In particular, the present invention relates to cooperative control among a plurality of storage devices constituting the storage system in which data is duplicated and stored.

"A Case for Redundant Array of Inexpensive Disks (RAID)" (Proceedings of ACM SIGMOD, 1988) defines a classification by a method of generating and storing redundant data in a disk array device consisting of a plurality of disk devices.

According to the above technique, by storing redundant data for two disk units, a failure occurs in one disk unit having a redundant configuration, and reading of the stored data of the disk unit or writing to the disk unit is prevented. Even if it is impossible, the disk array device which enables reading and writing of data using another disk device is called RAID1.

In addition, a technique for improving the processing performance of a disk control device by reducing the transfer processing of write data to a disk device redundant from the disk control device in the disk array device in the RAID array disk array device is disclosed in Japanese Patent Application Laid-Open No. 10-74129. 1 prior art).

According to the first conventional technique, at the time of data writing, the disk control apparatus transmits write data only to the first disk apparatus, and the write data is transmitted to the second disk apparatus without passing the write data through the disk control apparatus. The transfer process of write data from the disk control device to the second disk device becomes unnecessary.

Further, a method in which two disk devices are coupled to each other by a dedicated communication line and communicates with each other at the time of data writing and simultaneously write processing is disclosed in Japanese Patent Laid-Open No. 7-281959 (second prior art).

According to the second prior art, by transmitting write data to two disk devices at the same time, the transfer amount of write data on the interface can be realized in the same manner as in the case where no duplication is performed.

In the above-described first conventional technology, a disk array device is required which controls at least the individual disk devices constituting the disk array device and has a means for issuing a special write command to the first disk device. The problem is that it becomes expensive.

In addition, in the above conventional technology, it is not possible to reduce the transfer amount of write data on the interface, that is, it is not possible to reduce the interface load connecting the disk control device and the disk device, thereby improving performance degradation due to contention of the interface in the disk array device. There is a problem that cannot be done.

In the second conventional technology, the disk control device, which has been a problem in the first conventional technology, becomes unnecessary and further prevents an increase in the transfer amount of write data on the interface. You need a ferry.

In addition, performance due to a decrease in the utilization efficiency of the disk device in order to realize the transfer process of write data while synchronizing with two disk devices or confirming the progress of the operation of the secondary device by the main device. There is a problem that the performance is degraded due to a decrease or extra processing time for the data write processing of one disk device.

In addition, in the above-mentioned prior art, it cannot be said that the first / second or main / side switching control method and the disk device selection control method during data read processing are sufficiently considered.

An object of the present invention is to provide a data read / write request from a host computer in a RAID1 storage system composed of a plurality of storage devices, and to cooperate with a plurality of storage devices forming the storage system. It is to provide a control method of processing. This provides a storage system that solves the above problems.

The present invention can also be applied to RAID systems other than RAID1.

In order to achieve the above object, in the storage system including the RAID1 constituted by a plurality of storage devices, the following configuration is taken.

According to the present invention, a storage device included in the storage system can transmit means for receiving a request for information processing of the storage system from the host device, and the received request to a second storage device included in the plurality of storage devices. And means for executing information processing indicated by the received request when the received request is a request to be executed by the first storage device.

Further, each storage device constituting the storage system includes means for sharing a write data request to the storage system from at least an upper level device and a paired storage device for the shared write data request. The data transmission process and the means for synchronizing the transmission process of the status information with respect to the write data request may be provided.

<Example>

EMBODIMENT OF THE INVENTION Hereinafter, 1st Embodiment which concerns on this invention is described using drawing.

FIG. 1 shows a storage system 4 composed of two storage devices 1-a / b, and a plurality of host devices 2 connected to an arbitrary interface (SAN: Storage Area Network: 3) in FIG. It is a figure which shows one structural example of the connected information processing system.

In Fig. 1, a configuration is shown by taking a disk device as an example of the storage device 1, and the disk devices 1-a / b are disks each of which is a storage medium for holding data accessed by the host device 2, respectively. To perform interface protocol control between the unit 11, the buffer unit 12 that temporarily holds data transferred between the disk unit 11 and the host device 2, and the host device 2; An interface control unit 13, a disc control unit 14 for executing a recording / reproducing process for the disc unit 11, a buffer control unit 15 for controlling access to the buffer unit 12, and the respective portions It consists of a main control unit 10 for controlling the overall.

In addition, in FIG. 1, the disk device (1-a / b) holds duplicated data to form the RAID1 storage system.

In addition, in FIG. 1, the case where the memory device system is comprised by two disk devices is shown as an example, The structure of the memory device system of this invention may contain not only this but 3 or more memory devices.

EMBODIMENT OF THE INVENTION Hereinafter, 1st Embodiment which concerns on this invention is described using FIG. 2, FIG. 3, FIG. 4, FIG. 5, FIG. 6, FIG. In this embodiment, each storage device has a request (or receives), and it is determined whether each storage device performs information processing corresponding to the request in the storage device. Coordination control information 28 is used for the determination.

In the present embodiment, the access request from the host device 2 to the storage system 4 is shared by the plurality of storage devices 1 constituting the storage system 4, and in each storage device. By judging whether or not the access request is to be processed by oneself, it is unnecessary to realize a high-performance storage system in which a control device for controlling individual storage devices is unnecessary, low cost, and an optimum storage device is expected to be used. Make it possible.

2 is a diagram showing an example of the configuration of the disk device 1 constituting the storage device system 4.

In Fig. 2, the interface control unit 13 is similar to the receiving unit 20 which receives the information via the interface 3, the transmitting unit 21 which transmits the information, and information such as the received frame. When receiving an error detection or storing at least a portion of the frame information in the buffer unit 12, the receiving frame processing unit 22 for controlling the storage destination, etc., and the frame information when transmitting information such as a frame A transmission frame generation unit (23) for generating additional information such as header information to be configured is provided.

In addition, a retransmission judging unit (24) for determining whether to retransmit information, such as a frame, received by the receiving unit 20 to a subsequent device and generating a retransmission control signal, and the retransmission control signal And an output selector 25 which selects one of the information from the transmission frame generator 23 and the information received by the receiver 20 and outputs the information to the transmitter 21.

In addition, the buffer unit 12 stores a receive command storage unit 26 that holds an access request (command) received from the host device 2, and transmit / receive data storage that holds data transmitted / received between the host device 2. The part 27 is provided.

In addition, the main control unit 10 can process the access request from the host device 2 to the storage device 4 from the host device 2 to the subsequent device, and the access request itself. Cooperative control information 28 between the pair of disk devices, such as information for determining whether or not it is, is held. In addition, the detail of cooperative control information 28 is mentioned later using FIG.

The retransmission determining unit 24 performs retransmission control based on the cooperative control information 28 for the access request as the retransmission control of information such as a frame from the host apparatus 2 to the storage system 4. For the write data, retransmission (transmission) control is performed based on the header information or the like constituting the frame of the write data.

Hereinafter, read / write data transfer processing in the case where the individual disk device determines whether or not the access request is handled by the access request area on the condition of the access request from the host device to the storage system is performed. As an example, the operation will be described.

3 is a diagram showing the flow of processing in each device when a read data request from the host device 2 is issued to the storage device system 4 constituted of the disk devices 1-a / b. to be.

(1) Access request (command) reception and retransmission processing

The read data request issued to the storage system 4 from the host device 2 is first received as the frame information via the receiving unit 20 of the disk device 1-a. The disk device 1-a stores the read data request in the receive command storage 26 of the buffer 12 through the receive frame processor 22.

Further, the retransmission determining unit 24 of the disk device 1-a issues a request for access to the storage device system 4 from the host device 2 based on the cooperative control information 28 to the rear disk device ( By setting retransmission to 1-b), it is detected that the received frame information is an access request, and generates a retransmission control (instruction) signal for the output selector 25. As a result, the output selector 25 resends the read data request from the host device to the disk device 1-b of the subsequent stage based on the retransmission control (instruction) signal.

In the disk device 1-b, the read data request retransmitted by the disk device 1-a is received by the receive command storage part 26 of the buffer part 12 through the reception frame processing part 22. Store in The retransmission determining unit 24 of the disk apparatus 1-b is set not to resend the access request to the subsequent apparatus based on the cooperative control information 28.

(2) Analyzing access request (command) and executing access request (command)

The disk device 1-a / b analyzes the access request stored in the received command storage 26 and detects that the access request is a read data request.

Further, based on the lead request area information constituting the lead data request and the cooperative control information 28, it is determined whether the lead data request is to be processed by itself.

In the disc device determined to be processed by the disc controller 14, the disc controller 11 starts the disc read process from the disc unit 11 to the transmit / receive data storage unit 27 of the buffer unit 12, and the interface control unit. Through 13, the read data transmission process to the host apparatus 2 is performed. After the read data transmission, a series of read data request processing is completed by generating and transmitting status information on the read data request, and deleting the access request.

On the other hand, the disk device that has determined that it does not process, completes the processing for the read data request by deleting the access request.

In the present embodiment, the access request is transmitted to the disk devices 1-b via the disk devices 1-a, but the host device 2 may transmit the access request to each of the disk devices.

Fig. 4 shows the host device 2- in particular when a read data request from the host device 2-a / b is issued to the storage system 4 composed of the disk devices 1-a / b. The read data request when the read data request from a) is processed in the disk device 1-a, and the read data request from the upper device 2-b is processed in the disk device 1-b, and the read data. It is a figure which shows the flow of a transfer process.

As described above, the read data request from the host device 2-a / b is transmitted at least through the interface control unit 13 of the disk device 1-a. 12).

The disk device (1-a / b) determines whether or not it will process the received access request, and the disk read process from the disk unit 11 for the read data request determined to be processed by itself, and the higher level. Read data transmission processing and status generation transmission processing for the apparatus 2-a / b are also performed. Here, the read data transmitted by the disk device 1-a is transmitted to the host device 2-a through the interface control unit 13 of the disk device 1-b as in the prior art. is sent to a).

FIG. 5 is a diagram showing the flow of processing in each device when a write data request from the host device 2 is issued to the storage device system 4 constituted of the disk devices 1-a / b. .

(1) Access request (command) reception and retransmission processing

The same as described above with reference to FIG. 3 except that the target access request is not a read data request but a write data request. The write data request issued to the storage device system 4 from the host device 2 is first received as the frame information through the receiving unit 20 of the disk device 1-a.

The disk device 1-a stores the write data request through the reception frame processing unit 22 to the reception command storage unit 26 of the buffer unit 12.

Further, the retransmission determining unit 24 of the disk device 1-a issues a request for access to the storage device system 4 from the host device 2 based on the cooperative control information 28 to the rear disk device ( By setting retransmission to 1-b), it is detected that the received frame information is an access request, and generates a retransmission control (instruction) signal for the output selection section 25. As a result, the output selector 25 retransmits the write data request from the host device to the disk device 1-b of the subsequent stage based on the retransmission control (instruction) signal.

The disk device 1-b transmits the write data request resent by the disk device 1-a to the received command storage unit 26 of the buffer unit 12 via the reception frame processing unit 22. Store in The retransmission determining unit 24 of the disk apparatus 1-b is set so as not to resend the access request to the subsequent apparatus based on the cooperative control information 28.

(2) Analyzing access request (command) and executing access request (command)

The same as described above with reference to Fig. 3 except that the access request to be the target is not the read data request but the write data request. The disk device 1-a / b analyzes the access request stored in the received command storage 26, and detects that the access request is a write data request. Further, based on the write request area information constituting the write data request and the cooperative control information 28, it is determined whether or not the write data request is to be processed by itself.

The disk device that has determined that the processing is to secure the transmission / reception data storage unit 27 of the buffer unit 12 and transmits a write data transmission start request to the host device.

Further, the write data transmitted from the host device 2 that has received the write data transmission start request is stored in the transmission / reception data storage unit 27 secured in advance. Further, after the write data reception or the completion of the disk write processing for recording the received write data to the disc unit 11, the status information for the write data request is generated and transmitted, and the access request is deleted. This completes the write data request process.

On the other hand, in the disk apparatus which judges that it does not process, it updates the unupdated area management information mentioned later based on the write request area information which comprises the said access request, and then completes a write data request process by deleting the said access request. .

Fig. 6 shows the host device 2-a in particular when a write data request from the host device 2-a / b is issued to the storage system 4 constituted of the disk devices 1-a / b. The write data transfer start request in the case where the write data request from &quot; 1 &quot; is processed in the disk device 1-a, and the write data request from the host device 2-b is processed in the disk device 1-b, and It is a figure which shows the flow, such as a transfer process of write data.

As described above, the write data request from the host device 2-a / b is transmitted to at least the buffer unit of the disk device 1-a / b through the interface control unit 13 of the disk device 1-a. 12).

The disk device (1-a / b) determines whether or not to process the received access request by itself, and transmits a write data transfer start request to a higher-level device requesting the write data request determined by the disk device (1-a / b). In response to this, the write data in which transmission is started from the host apparatus is stored in the transmission / reception data storage unit 27.

Further, after the write data reception or after the disk write processing for recording the received write data to the disc unit 11, completion and generation of the status information for the write data request is performed.

Here, the transfer start request and the status information of the write data transmitted from the disk device 1-a to the host device 2-a are the same as those of the above-described read data transfer process. It is transmitted to the host device 2-a through the interface control unit 13 of b).

In addition, with respect to the write data transmitted from the host apparatus 2-a / b to the storage device system 4, retransmission (transmission) control based on header information or the like constituting a frame of the write data is executed. The retransmission determining unit 24 of the interface control unit 13 of 1-a / b executes. That is, the write data from the host device 2-a is based on the header information constituting the frame of the write data in the retransmission determining unit 24 of the interface control unit 13 of the disk device 1-a. As a result of the retransmission (transmission) determination control, it is stored in the buffer unit 12 of the disk device 1-a so as not to be transferred to the next stage.

Regarding the write data from the host apparatus 2-b, the retransmission determining unit 24 of the interface control unit 13 of the disk apparatus 1-a is based on the header information constituting the frame of the write data or the like. As a result of the retransmission (transmission) determination control, the transmission to the next stage is performed, and the retransmission determination unit 24 of the interface control unit 13 of the disk apparatus 1-b constitutes the frame of the write data. As a result of the retransmission (transfer) determination control based on the header information or the like, it is stored in the buffer unit 12 of the disk apparatus 1-b so as not to be transmitted to the next stage.

As described above, in the present embodiment, in a storage system composed of a plurality of storage devices, individual storage devices constituting the storage system request an access request to the storage device system from a higher level device. Means shared by the device, means for determining whether to process the shared access request by the device, and also write data transmitted from the host device to the storage system to the storage device determined to be processed, at least. By providing the means, a high-performance storage system capable of realizing a low cost and optimum use efficiency of the storage device without the need for a control device for controlling the individual storage devices can be realized.

In addition, in the present embodiment, inconsistency in the data stored in the disk devices 1-a / b occurs temporarily in accordance with the write data request from the host device 2. Thus, an example of the write data update process according to the present embodiment will be described below.

FIG. 7 is a diagram showing an example of the configuration of unupdated area management information 30 that is subjected to update processing in the disk device which determines that the write data request from the host device described above is not processed.

The unupdated area management information 30 has at least information on the unupdated area. In FIG. 7, the case where the access request from the host device to the storage system is composed of access start address information and access size information is taken as an example. For example, it is set as the structure which has unupdated start address information and unupdated size information, and also contains update information, such as update time information regarding an unupdated area.

The disk device that receives the write data request from the host device and judges that it does not process itself, updates the unupdated area management information 30 on the basis of the write request area information constituting the write data request. Management of the update area and subsequent update process of the unupdated area are made possible.

FIG. 8 shows a flow of an update request and an update data transfer process in the case of performing an update process of an unupdated area where a discrepancy has occurred based on the unupdated area management information 30 of the disk device 1-b. Drawing.

In FIG. 8, the disk device 1-b issues an update request to the paired disk devices 1-a based on its unupdated area management information 30. In FIG. In the disk device 1-a which has received the update request, a disk read from the disk unit 11 is performed if necessary, and the data to be requested is transmitted to the disk device 1-b. In the disk device 1-b, the data received from the disk device 1-a is stored in the disk unit 11, and the unupdated area information required for the processing is deleted from the unupdated area management information 30. The update process of the unupdated area is completed.

In addition, in the above-described embodiment, the disk device that receives the write data request from the host device and judges that it does not process itself performs the management of the unupdated area in which a mismatch occurs and the subsequent update process of the unupdated area. However, not only this but also the unupdated area in which inconsistency occurs, for example, by the disk device determined to be processed by itself updating the unupdated area management information 30 based on the write request area information constituting the write data request. It is also possible to carry out the management and the update processing of the unupdated area afterwards.

In addition, the occasion which starts the update process of an unupdated area is arbitrary. For example, it is possible to start based on the number of entries registered in the unupdated area management information 30, or the total unupdated area size.

Hereinafter, 2nd Embodiment which concerns on this invention is described using FIG. 9, FIG. 10, FIG. 11, FIG. 12, and FIG. In this embodiment, it is determined whether or not the received disk device should first process the received access request. As a result of the discrimination, it is characterized in that the access request is transmitted to another disk device unless it is to be processed by the disk device. That is, in the present embodiment, the access request for the storage system from the host device is received by one storage device constituting the storage system, and the storage device for which the access request is to be processed in the storage device is determined and necessary. In this case, by transmitting the access request from the host device to the storage device to be processed, a control device for controlling the individual storage devices is unnecessary, low cost, and high performance that is expected to use the optimum storage device. The storage device system can be realized.

9 is a diagram showing an example of the configuration of a disk device 1 constituting the storage device system 4. In FIG. 9, the same configuration as in FIG. 2 is provided except that a command transfer control unit 29 is newly provided to control whether or not to transmit the access request received from the host device 2 to the paired disk devices. It is.

As the cooperative control information 28, information for performing control of determining whether or not to transmit the access request received from the host device 2 to a paired disk device is held. In addition, the retransmission determining unit 24 performs retransmission (transmission) based on header information or the like constituting a frame of the write data as retransmission control of information such as a frame from the host device 2 to the storage system 4. ) To perform control.

Hereinafter, the read / write data transfer process in the case where the individual disk device determines whether or not the access request should be processed by the host device with respect to the access request area from the host device to the access request area is performed. For example, the operation will be described.

FIG. 10 shows the flow of processing in each device when a read data request from the host device 2 is issued to the storage device system 4 constituted of the disk devices 1-a / b. Drawing. In addition, in the following description, the disk device 1-a receives an access request from the host device 2, determines a device to process the access request, and if necessary, sends the access request to the disk device ( Transmission control is performed for transmission in 1-b).

(1) Receive access request (command)

The lead data request issued to the storage system 4 from the host device 2 is frame information. First, a buffer unit (1) is provided through the receiving unit 20 and the receiving frame processing unit 22 of the disk apparatus 1-a. 12 is stored in the received command storage unit 26 of FIG.

(2) access request (command) analysis and access request (command) transfer processing

The disk device 1-a analyzes the access request stored in the reception command storage 26, and detects that the access request is a read data request. Further, based on the lead request area information constituting the lead data request and the cooperative control information 28, it is determined whether the lead data request is to be processed by itself.

In the above determination, when it determines that it processes, the access request is processed in the disk device 1-a. If it determines that it does not process, the disk device 1-a transmits a read data request from the host device 2 to the disk device 1-b.

In the disk apparatus 1-b, the read command request transmitted from the disk apparatus 1-a is received by the receiving unit storage unit 26 of the buffer unit 12 via the receiving unit 20 and the receiving frame processing unit 22. Store in The disk device 1-b also analyzes the access request stored in the received command storage 26 and detects that the access request is a read data request.

(3) Execution of access request (command)

In the disk device 1-a or the disk device 1-b in which the access request is transmitted from the disk device 1-a, which has been determined to be processed by the disk device 1-a, the disk control unit 14 uses the disk control unit 14 based on the read data request. The disc read process from the disc unit 11 to the transmission / reception data storage unit 27 of the buffer unit 12 is started. Further, the interface control unit 13 performs read data transmission processing to the host apparatus 2. After the read data transmission, a series of read data request processes are completed by generating and transmitting status information for the read data request and deleting the access request. If the disk device 1-a determines that the disk device 1-a does not process itself, the process of requesting the read data request is completed by deleting the read data request.

Fig. 11 shows the host device 2-a, in particular, when a read data request from the host device 2-a / b is issued to the storage system 4 composed of the disk devices 1-a / b. Read data request from the disk device 1-a, and the read data request from the host device 2-b in the disk device 1-b, and the transfer of the read data. It is a figure which shows the flow of a process.

As described above, the read data request from the host device 2-a is received by the disk device 1-a and is not transmitted to the disk device 1-b. The read data request from the host device 2-b is received by the disk device 1-b by the transfer control process of the disk device 1-a described above.

In the disk device 1-a / b, the disk read process from the disk unit 11, the read data transmission process to the host device 2-a / b, and the status generation transmission to the received read data request. Perform the process.

Here, since the read data transmitted by the disk device 1-a is for the host device 2-a, the host device 2 is connected through the interface control unit 13 of the disk device 1-b as in the related art. sent to -a).

FIG. 12 is a diagram showing the flow of processing in each device when a write data request from the host device 2 is issued to the storage device system 4 constituted of the disk devices 1-a / b. .

(1) Receive access request (command)

The write data request issued to the storage system 4 from the host device 2 is frame information. First, a buffer is provided through the receiving unit 20 and the receiving frame processing unit 22 of the disk device 1-a. The received command storage unit 26 of the unit 12 is stored.

(2) access request (command) analysis and access request (command) transfer processing

The disk device 1-a analyzes the access request stored in the reception command storage 26, and detects that the access request is a write data request. Further, based on the write request area information constituting the write data request and the cooperative control information 28, it is determined whether or not the write data request is to be processed by itself.

If it is determined by the determination that the processing is itself, the access request is processed by the disk device 1-a. If it determines that it does not process, the disk device 1-a transmits a write data request from the host device 2 to the disk device 1-b.

The disk device 1-b stores the write data request transmitted by the disk device 1-a via the receiving unit 20 and the receiving frame processing unit 22 in the receiving command of the buffer unit 12. Stored in section 26. The disk device 1-b also analyzes the access request stored in the received command storage 26 and detects that the access request is a write data request.

(3) Execution of access request (command)

In the disk device 1-a or the disk device 1-b in which the access request has been transmitted from the disk device 1-a, the buffer unit 12 transmits and receives based on the write data request. The data storage unit 27 is secured and a write data transmission start request is sent to the host device. Further, the write data transmitted from the host apparatus having received the write data transfer start request is stored in the previously-transmitted transmit / receive data storage unit 27.

Further, after the write data reception or the completion of the disk write process for recording the received write data to the disc unit 11, the status information for the write data request is generated and transmitted.

(4) Access request (command) completion processing

In the disk apparatus 1-b, after generating and transmitting the status information, the write request request process is completed by deleting the access request. Further, the disk device 1-a determines that the status information is not generated after the generation or transmission of the status information, or the disk device 1-a transmits a write data request from the host device 2 to the disk device 1-b. Subsequently, based on the write request area information constituting the write data request, the unupdated area management information described later is updated. In addition, by deleting the access request, a series of write data request processing is completed.

In the present embodiment, particularly in the case where a write data request from the host device 2-a / b is issued to the storage device system 4 comprised of the disk devices 1-a / b, Start of transfer of write data when the write data request from (2-a) is processed by the disk device 1-a and the write data request from the host device 2-b is processed by the disk device 1-b. Since the request, the flow of the write data transfer process, and the like are the same as those in FIG.

However, as described above, the access request from the upper apparatuses 2-a / b is stored at least in the buffer unit 12 of the disk apparatuses 1-a, and the disk apparatuses 1-a make their own. If it is determined that no processing is performed, it is transmitted to the disk device 1-b.

As described above, in the present embodiment, in a storage system composed of a plurality of storage devices, at least one storage device constituting the storage system receives an access request from the host device to the storage device. In addition, in the storage device, by determining the storage device to which the access request should be processed, and if necessary, by providing means for transferring the access request to the storage system from the host device to the storage device to be processed. A high-performance storage system can be realized that is inexpensive because the control device for controlling the individual storage devices is unnecessary, and that the use efficiency of the optimum storage device is expected.

In addition, similarly to the above-described embodiment, in this embodiment, there is a discrepancy between the data stored in the disk devices 1-a / b in accordance with the write data request from the host device 2. Therefore, an example of the update processing of write data in the present embodiment will be described below.

In the present embodiment, the unupdated area management information is added to the above-described unupdated area management information 30 by adding information on whether or not the data to be stored is new (hereinafter referred to as update type information). Based on the write request area information, which is held from -a) and constitutes write data requests from the host device in the disk device 1-a, the unupdated area information constituting the unupdated area management information and the update type. By updating and managing the information, management of the unupdated area in which inconsistencies have occurred and subsequent update process of the unupdated area are realized.

Fig. 13 is a diagram showing a flow of an update request and an update data transfer process in the case of performing an update process of an unupdated area where a discrepancy has occurred based on the unupdated area management information of the disk apparatus 1-a. to be. 13, the case where the disk device 1-a updates itself using the latest data stored in the disk device 1-b is taken as an example.

The disk device 1-a requests an update request (latest data transmission request) for the paired disk devices 1-b based on the unupdated area management information composed of at least unupdated area information and update type information. ). In the disk device 1-b that has received the update request, the disk device 11 reads the disk from the disk unit 11 if necessary, and transmits data to be requested to the disk device 1-a. In the disk device 1-a, the data received from the disk device 1-b is stored in the disk unit 11, and the unupdated area information required for the processing is deleted from the unupdated area management information 30. This completes the update processing of the unupdated area.

When the disk device 1-a updates the disk device 1-b using the latest data stored by the disk device 1-a, the disk device 1-a is paired based on the unupdated area management information. An update request (latest data reception request) is issued to the disk device 1-b which constitutes the next, and the data to be updated is subsequently transmitted. In the disk device 1-b that has received the update request and the data to be updated, the disk unit 11 stores the data received from the disk device 1-a. In addition, the disk device 1-a completes the update process of the unupdated area by deleting, from the unupdated area management information 30, the unupdated area information required for the processing after the update target data is transmitted.

In the above embodiment, the operation has been described taking as an example the case where the host device determines whether or not the access request is to be processed by itself, subject to the access request area constituting the access request to the storage system. It is not limited to.

For example, when the storage device manages data stored in units of objects (files), the access request itself is subject to object (file) information constituting an access request from the host device to the storage system. It is possible to determine whether or not to perform the processing, and by using object (file) information for the unupdated area information held at least as the unupdated area management information, it is possible to realize the above-described unupdated area update processing. .

Further, in the above embodiment, in a RAID1 storage system composed of a plurality of storage devices, each storage device constituting the storage system or at least one storage device accesses the storage system from an upper device. By providing means for determining whether or not to process the request itself, the control device for controlling the individual storage devices becomes unnecessary. In addition, the above-described high performance storage system is expected to realize the optimum storage efficiency.

However, in order to realize the optimum use efficiency of the storage device, although it is temporary depending on the write data request from the host device, inconsistency occurs in the data of the redundant disk device, and the write data update process may be required. . In addition, in the update process of the write data, it is not possible to reduce the load of the interface 3 in order to perform the transfer process of write data through the interface 3.

Therefore, the access request, especially the write data request, to the storage system 4 from the host device 2 is shared by the plurality of storage devices 1 constituting the storage system 4 and further paired. An embodiment in which high performance is achieved by reducing the interface load by synchronously processing the write data request in a plurality of storage devices will be described below.

Hereinafter, 3rd Embodiment which concerns on this invention is described using FIG. 14, FIG.

In the present embodiment, a plurality of memories that share and write a pair of write data requests from the host device 2 to the storage system 4 to the plurality of storage devices 1 constituting the storage system 4. By carrying out the write data transfer processing in synchronization with the device, the control device for controlling the individual storage devices is unnecessary, thereby realizing the above-mentioned high-performance storage device system which is low cost and reduces the interface load.

In addition, the structure of the memory | storage device (disk apparatus) which comprises the memory | storage system of this embodiment is sent by the retransmission determination control part 24 to the retransmission of the information, such as a frame, with respect to the memory | storage system 4 from the host apparatus 2. As the control, since it is the same as in FIG. 2 except for the retransmission control based on the said cooperative control information 28 also about write data, it abbreviate | omits.

The read data request from the host device to the storage system is not the object of the present embodiment, and any one of the above-described embodiments is adopted, and the description thereof is omitted.

The operation will be described below by taking the write data transfer process when the write data transfer is performed in synchronization with the write data request from the host device to the storage system in two pairs of storage devices.

FIG. 14 is a diagram showing the flow of processing in each device when a write data request from the host device 2 is issued to the storage device system 4 constituted of the disk devices 1-a / b. to be.

FIG. 15 shows a write data request when a write data request from the host device 2 is issued and a write start transfer request for the storage device system 4 constituted of the disk devices 1-a / b. And a flow of the write data transfer process and the like.

(1) Access request (command) reception and retransmission processing

The write data request issued to the storage device system 4 from the host device 2 is first received as the frame information through the receiving unit 20 of the disk device 1-a.

The disk device 1-a stores the write data request in the receive command storage 26 of the buffer 12 through the receive frame processor 22.

In addition, the retransmission determining unit 24 of the disk device 1-a issues a request for access to the storage device system 4 from a higher level device based on the cooperative control information 28 to the rear disk device 1-b. ), It is detected that the received frame information is an access request, and generates a retransmission control (instruction) signal for the output selector 25. As a result, the output selector 25 retransmits the write data request from the host device to the disk device 1-b of the subsequent stage based on the retransmission control (instruction) signal.

In the disk device 1-b, the write data request resent by the disk device 1-a is transmitted through the reception frame processing part 22 to the reception command storage part 26 of the buffer part 12. ). The retransmission determining unit 24 of the disk apparatus 1-b is set so as not to resend the access request to the subsequent apparatus based on the cooperative control information 28.

(2) Analyzing access request (command) and executing access request (command)

The disk device 1-a / b analyzes the access request stored in the reception command storage 26, detects that the access request is a write data request, and transmits / receives the data transmission / reception of the buffer unit 12 ( 27). The disk device 1-a transmits a write data transfer start request to the disk device 1-b on the basis of securing its own transmission / reception data storage unit 27. On the other hand, in the disk device 1-b, the host device 2 is secured on the basis of securing its own transmission / reception data storage unit 27 and reception of a transfer start request for write data from the disk device 1-a. Transmits a write data transfer start request.

The host device 2 starts the transmission of the write data in response to receiving the write start request of the write data.

The write data transmitted from the host device 2 is first received through the receiving unit 20 of the disk device 1-a. The disk device 1-a stores the write data in the transmission / reception data storage unit 27 of the buffer unit 12 through the reception frame processing unit 22.

In addition, the retransmission determining unit 24 of the disk device 1-a reads write data from the host device 2 to the storage device system 4 from the host device 2 based on the cooperative control information 28. By setting -b) to retransmit, it is detected that the received frame information is write data and generates a retransmission control (instruction) signal for the output selection unit 25. The output selector 25 retransmits the write data from the host device to the disk device 1-b at the next stage based on the retransmission control (instruction) signal.

The disk device 1-b stores the write data retransmitted by the disk device 1-a in the transmission / reception data storage unit 27 of the buffer unit 12 via the reception frame processing unit 22. In addition, the retransmission determining unit 24 of the disk apparatus 1-b is set so as not to retransmit the write data to the subsequent apparatus based on the cooperative control information 28.

In the disk device 1-a, after the write data is received or after the completion of the disk write process of recording the received write data to the disk unit 11, the status of the write data request is given to the disk device 1-b. Generate and send information. Furthermore, the series of write data request processings is completed by deleting the write data request.

In the disk device 1-b, after receiving the write data or completing the disk write process of recording the received write data to the disk unit 11, and receiving the status information from the disk device 1, the host device For (2), status information on the write data request is generated and transmitted, and the write data request processing is completed by deleting the write data request.

As described above, in the present embodiment, in the RAID1 storage system composed of a plurality of storage devices, each storage device constituting the storage system requests write data requests from the higher-level device to the storage device system. By means of sharing, means for synchronizing the transfer process of write data and the transfer process of status information with respect to the write data request between the plurality of storage apparatuses paired with the shared write data request, The control device for controlling the individual storage devices is unnecessary, low cost and data update processing between the storage devices are unnecessary, and the above-mentioned high performance storage device system that reduces the interface load can be realized.

In the above embodiment, although the storage system of RAID1 comprised with only the some memory device which applied this invention was mentioned, it is not limited to this. For example, even in a disk array apparatus comprising a conventional disk array apparatus, that is, a plurality of disk apparatuses and a control apparatus for controlling these disk apparatuses, the configuration of the control apparatus becomes easier by employing the disk apparatus to which the present invention is applied. As a result, the price can be lowered.

In addition, an example of the determination process of whether or not the above-mentioned access request from the host apparatus itself processes is described below.

16 is a diagram illustrating an example of information constituting the cooperative control information 28.

In Fig. 16, as the information constituting the cooperative control information 28, the device identification information (Storage Subsystem ID) of the storage system 4 and the device identification information (Storage Device ID) unique to each storage device itself. ), Device identification information (Cooperation Storage ID) inherent in the storage device which performs cooperative control with itself, cooperative processing mode information, and cooperative area size information and active area start as the area management information to be cooperatively processed. The address information and the active area size information are held.

In addition, an example of setting in the storage device 1-a / b is shown for each of the above information, and the device identification information of the storage device system 4 is ID_0 for each of the storage devices 1-a / b, and each of the pieces of information. Device identification information unique to the storage device itself is ID_1 for the storage device 1-a, ID_2 for the storage device 1-b, and the device identification information unique to the storage device for cooperative control with itself is stored in the storage device 1. ID_2 is set in -a) and ID_1 is set in the storage device 1-b, the cooperative processing mode information is set to resend the command in the storage device 1-a, and the command receipt is set in the storage device 1-b. The cooperative area size information is N, the active area start address information is 0 in the storage device 1-a, the N / 2 in the storage device 1-b, and the active area size information is the storage device 1-a / b. ) N / 2 are all set.

In addition, the device identification information of the storage system 4, the device identification information unique to each storage device itself, the device identification information unique to the storage device which performs cooperative control with itself, and the cooperative processing mode information Cooperative control information used in the frame data retransmission process determination described later.

FIG. 17 is a diagram showing an example of the configuration of access request information specified in FCP (Fibre Channel Protocol) as access request information from the host apparatus. FIG. In Fig. 17, the access request information includes start address information (Logical Block Address in Fig. 17) of the access request target area and access request target area size information (Transfer Length).

The storage device 1 that has received the access request information from the host device, access target object of active area start address information of the cooperative control information held by the host device, active area size information, and access request information from the host device. Based on the start address information of the area and the access request target area size information, it is determined whether or not the access request is to be processed by itself, that is, whether it is an access request to its own active area, Processing is performed only for the access request.

For example, when the access request target area from the host device is N / 2 or less, when the storage device # 1 is N / 2 or more, the storage device # 2 processes the access request from the host device. In addition, when the access request target area from the host device corresponds to the active areas of the storage device # 1 and the storage device # 2, the storage device # 1 and the storage device # 2 respectively process access requests to their own active area. .

In the above-described embodiment, the access request is based on the address information and the size information and the address information constituting the access request information from the host apparatus as the management information of the cooperative processing target area. Although the case of discriminating whether or not the processing is performed by oneself has been described as an example, not only this, but also the logical unit number shown in FIG. 17 (FCP_LUN or Logical Unit Number in FIG. 17) may be used as a condition. In this case, the storage device 1 holds active logical unit number information as management information about a cooperative processing target area constituting the cooperative control information 28.

In addition, when the access request information from the host device holds an object name such as a file name, the storage device 1 is an active object list as management information about the cooperative processing target area constituting the cooperative control information 28. By retaining the information, it is also possible to determine whether or not to process the access request from the host apparatus on an object basis.

In the second embodiment described above, when the access request target area from the host device corresponds to the disk areas 1-a and the active areas of the disk devices 1-b, The access request information from the host device is processed into an access request for the active area of the disk device 1-b, and transmitted to the disk device 1-b, whereby the disk device 1-a / b is used. It is possible to realize cooperative control processing.

Alternatively, the above-mentioned cooperative control information (active area start address information, active area size information, etc.) is held in the disk device 1-b, and based on the access request target area information of the access request information from the host device. By determining whether or not the access request is to be processed by itself, i.e. whether it is an access request to its own active area, and processing the access request to its own active area. 1-a does not process the access request information from the host device into an access request for the active area of the disk device 1-b and transmits it to the disk device 1-b. It is possible to realize the cooperative control process in a / b).

In addition, an example of the determination process of whether or not to retransmit information, such as frame data transmitted between the above-mentioned upper apparatus and a storage system, with respect to a later apparatus is demonstrated below.

FIG. 18 is a diagram showing an example of the configuration of frame header information defined in FC-PH (Fibre Channel PHYSICAL AND SIGNALING INTERFACE) as information transmitted between an upper device and a storage system.

In FIG. 18, the frame header information includes information on a transmission destination / transmitter of frame data (D_ID: Destination_Identifier, S_ID: Source_Identifier), frame data type information (R_CTL: Routing Control, TYPE: Data structure type), and frame data. It has fields of associated exchange identification information (RX_ID: Responder Exchange_Identifier, OX_ID: Originator Exchange_Identifier).

First, in the case of frame data (lead data, transfer start request of write data, status information, etc.) transmitted from the storage device system 4 (the disk devices 1-a / b) to a higher level device, the frame Identification information of the host apparatus is set as the transmission destination information D_ID of the frame data of the header information, and the disk apparatus which has received the frame data retransmits.

Therefore, hereinafter, frame data (access to which the identification information of the storage system 4 is set as the transmission destination information D_ID transmitted from the host device to the storage system 4, that is, the frame data of the frame header information) is set. An example of retransmission determination processing will be described for a request, write data).

(1) access request

When the retransmission determining unit 24 receives the frame data (access request) for the storage system 4 from the host device, the frame data is accessed from the frame data type information R_CTL and TYPE of the frame header information. On the basis of the cooperative processing mode setting information (command retransmission or command receipt) constituting the cooperative control information 28 for detecting that the request is requested, it is determined whether to resend the received access request to the subsequent device. By determining whether or not, retransmission processing of frame data (access request) is realized.

(2) write data

When the retransmission determining unit 24 receives the frame data (write data) for the storage system 4 from the host device, the frame data is written from the frame data type information R_CTL and TYPE of the frame header information. Data and detects whether or not the source information (S-ID) of the frame data of the frame header information and the exchange identification information (RX_ID, OX_ID) match those registered in the write processing management information described later. If it matches, the write data is received, and if there is a mismatch, resending the frame data (write data) is realized by resending the device to a later stage.

Fig. 19 shows an example of the configuration of the write processing management information. The exchange identification information RX_ID and OX_ID of the write processing being executed (unfinished) and the identification information (Host ID) of the access request issuer host device are shown. Is an example of maintaining associating with the write processing management information. In Fig. 19, for example, the write data frame transmitted to the storage system 4 from the same host device (Host ID (S_ID) = ID_3) is the exchange identification information (RX_ID) constituting the frame header. and / or OX_ID) detect whether or not it is a write data frame to be received by detecting whether or not OX_ID matches that registered in the write processing management information.

In addition, as the minimum configuration example of the write processing management information, it is also possible to configure the exchange identification information OX_ID of the running (unprocessed) write processing and the identification information (Host ID) of the access request issuer host device. The OX_ID is a value set by the host device and does not guarantee independence. On the other hand, since the RX_ID can be set to any value on the storage device side, it is preferable to keep it as the write processing management information, and to set different values between the storage devices to be subjected to cooperative control. It is preferable.

Next, a fourth embodiment of the present invention will be described.

In the fourth embodiment, control information (cooperative control information) for individual storage devices constituting a storage device subsystem for coordinating between a plurality of storage devices to process an access request from a higher level device, and the storage device sub address from the higher level device, is provided. Based on the access request information for the system, it is determined whether or not the access request should be handled by itself.

In the fifth embodiment, the storage device of the fourth embodiment adds and maintains operation state information of the storage device other than the cooperative processing, so that even when a failure occurs in the storage device constituting the storage device subsystem, the host device To implement access request processing for the storage subsystem.

In the sixth embodiment, the storage device of the above-described embodiment further includes a plurality of interface control units, and transfers the access request information from the host device and the user data transferred between the storage devices constituting the storage subsystem. The processing is realized by using another interface (transmission path).

Fourth Embodiment (Normal Operation)

Hereinafter, 4th Embodiment which concerns on this invention is described using FIGS. 20-14.

FIG. 20 is a diagram showing an example of an information processing system composed of a storage device subsystem 4 composed of a plurality of (N, N> 1) storage devices 1 and an upper level device 2; In this case, the individual storage devices 1 constituting the storage device subsystem 4 are directly connected to the host device 2 through a cyclic (hereinafter, referred to as loop) interface 3.

In addition, in FIG. 20, the structure is shown taking the disk device as the storage device 1 as an example, and each disk device 1 is a disk part which is a storage medium which holds the data which said host device 2 accesses ( 11), a buffer unit 12 for temporarily holding data transmitted between the disk unit 11 and the host device 2, and an interface controller for performing interface protocol control between the host device 2 ( 13), a disc controller 14 for executing a recording / reproducing process for the disc unit 11, a buffer controller 15 for controlling access to the buffer unit 12, and data writing from the host apparatus. An update information generation unit 16 for generating update information by an exclusive-OR operation from the write data (new data) received on the basis of the request and the update target data (old data) stored in the disc section; It is a structure provided with the main control part 10 which collectively controls a site | part.

FIG. 21 is a diagram showing an example of the configuration of a disk device 1 constituting the storage device system 4.

In FIG. 21, the interface control unit 13 is similar to the receiving unit 20 which receives the information via the interface 3, similarly to the transmitting unit 21 which transmits the information, and information such as the received frame. The frame information at the time of transmitting information such as a frame and the reception frame processing unit 22 which performs control of a storage destination or the like when the error detection or the storage of at least a part of the frame information is stored in the buffer unit 12. A transmission frame generating unit 23 for generating additional information such as header information constituting the?, And whether information such as a frame received by the receiving unit 20 is to be retransmitted to a subsequent device, or One of the frame reception control section 24 which controls whether or not the user receives the data (retransmission control), the information retransmitted from the frame reception control section 24, and the information from the transmission frame generation section 23. And a frame transmission control unit 25 for outputting to the transmission unit 21.

In addition, the buffer unit 12 stores a receive command storage unit 26 that holds an access request (command) received from the host device 2, and transmit / receive data storage that holds data transmitted / received between the host device 2. And an update information storage unit 28 for holding update information generated by the update information generation unit 16.

In addition, the main control unit 10 carries out its own processing corresponding to the access request based on the cooperative control information 29 described later and the access request information from the host device 2 to the storage system 4. The cooperative process control part 30 to control is provided.

The retransmission control in the frame receiving control section 24 is performed based on the control information from the cooperative processing control section 30, header information constituting the frame, and the like.

FIG. 22 is a diagram showing an example of the configuration of the cooperative control information 29 and the settings in the individual storage devices. In Fig. 22, the cooperative control information 29 includes device identification information (shared ID) of the storage device subsystem 4, device identification information (unique ID) unique to each storage device itself, and the circulation. The position map information as the relative position information in the type interface 3, and the command transmission mode information specifying a mode of whether or not to retransmit the access request frame received from the front end to the rear end device.

In addition, RAID level information, redundant data management size, and the storage subsystem 4 constitute the number of data disks / redundant data disks as configuration information for management information of the RAID configuration. It consists of information, and arbitrary settings are made for each storage device.

FIG. 23 is a diagram illustrating a configuration of frame information in a Fiber Channel Arbitrated Loop (FC-AL) as an example of the circuit type interface 3.

In FIG. 23, frame information includes an SOF (Start of Frame) indicating the start of frame information, header information including type information of frame information, transmission destination / original information, and the like, and a payload (user) transmitted between devices. Data), the header information, a cyclic redundancy check (CRC) that is redundant data for the payload, and an end of frame (EOF) indicating the end of the frame information.

24 is a diagram showing an example of the header information, and is composed of R_CTL / TYPE, which is the type information of the frame information, and D_ID / S-ID, which is the source / source information of the frame information.

Fig. 25 is a diagram showing an example of access request information as the payload, and is a logical unit number (FCP LUN), address information (Logical Block Address), and transfer length information (Transfer Length), which are information for designating an object of an access request. And an operation code, which is the type information of the access request.

Hereinafter, an operation in which the storage devices 1-1 to N cooperatively process an access request from the host device 2 to the storage device subsystem 4 will be described.

(1) Access request (command) reception and retransmission processing

FIG. 26 is a diagram showing an example of processing in each storage device when an access request from the host device 2 is issued to the storage device subsystem 4 constituted of the storage devices 1-1 to N. FIG. to be.

First, an access request (Command) issued from the host device 2 to the storage device subsystem 4 is received as the frame information through the receiving unit 20 of the storage device 1-1.

In the storage device 1-1, the access request is stored in the reception command storage unit 26 of the buffer unit 12 via the frame reception control unit 24 and the reception frame processing unit 22.

Further, the frame receiving control section 24 of the storage device 1-1 carries out the access request to the storage device subsystem 4 from the host device 2 with the command transfer code of the cooperative control information 29. By setting the device to retransmit (receive & retransmit), it is detected from the header information of the received frame information that the frame information is an access request, and is transmitted from the host apparatus via the frame transmission control section 25. The access request is resent to the subsequent device (receive & resend).

Also in the subsequent storage devices 1-2 (N-1), the same command reception & retransmission process as described above is performed.

In the storage device 1-N of the last stage, the command transmission code of the cooperative control information 29 is not retransmitted from the higher-level device 2 to the storage device of the access request to the storage device subsystem 4 from the higher-level device. By setting (reception only), the received command storage unit of the buffer unit 12 detects that the frame information is an access request from the header information of the received frame information, and does not retransmit it to a subsequent device. Only storage to (26) is carried out.

By the above-described access request (command) reception and retransmission processing, the access requests from the host apparatus 2 are shared without starting the transfer process of the new access request in each of the storage devices 1-1 to N performing the cooperative process. It becomes possible.

(2-1) Read processing (processing is realized in one memory device)

FIG. 27 is a diagram showing an example of data transfer processing when the access request from the host device 2 is a read data request and can be processed by one storage device. In FIG. 27, the storage device 1-2 is shown. Assumes data transmission processing (Data2 transmission).

Each of the storage devices 1-1 to N determines whether or not the access request is processed by itself based on the access request information stored in the received command storage 26 and the cooperative control information 29. .

In the storage device 1-2 determined to process itself, the transmission / reception data storage unit 27-2 of the buffer unit 12-2 from the disc unit 11-2 via the disc control unit 14-2. The disc read process is started, and the read data transmission process is performed to the host device 2 via the interface control unit 13-2. After the read data transmission, a series of read data request processings are completed by generating and transmitting status information on the access request and deleting the access request.

On the other hand, the disk device that has determined that it does not process, completes the processing for the read data request by deleting the access request.

Further, in the storage devices 1-3 to N located at the rear end of the storage device 1-2, the frame information (lead data, status) transmitted from the storage device 1-2 is retransmitted from the header information. Only processing is performed.

(2-2) Read processing (processing is realized by multiple memory devices)

FIG. 28 is a diagram showing an example of data transfer processing in each storage device when the access request from the host device 2 is a read data request and processing in a plurality of storage devices is required. In FIG. It is assumed that the device 1-1 / 2 performs data transmission processing (Data1 / Data2 transmission).

FIG. 29 is a diagram showing an example of the configuration of data frame (payload) information when data transfer processing is required in a plurality of storage devices, whereas the conventional data frame (payload) is composed of only user data. The data is divided into user data segments 0 to n which are divided for each storage device to be processed, and a flag (Data Valid bit) indicating whether or not the data of the user data segment is valid.

In addition, in the following description, based on the access request information and the cooperative control information 29, the storage device for processing the access request is located at the foremost (receives the information from the host device fastest). Although the case where the storage device permits the transmission start of read data / status information is described as an example, not only this, but also the storage device which starts the transmission of read data by any algorithm can be determined.

As described above, each of the storage devices 1-1 to N decides whether to process the access request based on the access request information stored in the reception command storage 26 and the cooperative control information 29. The disk device that judges whether or not the processor has determined that the processor does not process the data and completes the processing for the read data request by deleting the access request.

In the memory device 1-1 / 2 that has determined to process itself, the transmission / reception of the buffer unit 12-1 / 2 from the disk unit 11-1 / 2 through the disk control unit 14-1 / 2 is performed. The disc read process for the data storage unit 27-1 / 2 is started.

In the storage device 1-1, which has been permitted to start the transmission of the read data, the user data is sent to the user data segment 0, for example, as the read data transmission process to the host device 2. In the data segment 1, dummy data is stored, and the data frame information transmission process in which "valid" is set to a flag corresponding to the user data segment 0 and "invalid" is set to a flag corresponding to another user data segment is performed. And after generating and transmitting the status information according to the processing, the series of read data request processings is completed by deleting the access request.

In addition, in the storage device 1-2 where the start of the transmission of the read data is prohibited, the read received from the storage device 1-1, for example, is a process of transmitting the read data to the host device 2. The lead data is stored in the user data segment 1 of the data frame, and the data frame information is transmitted with the " valid " set to a flag corresponding to the user data segment 1, and the status according to the process is performed. After generating and transmitting new status information based on the information and the status information received from the preceding apparatus, a series of read data request processings are completed by deleting the access request.

In addition, in a storage device of the last stage that performs the data transmission process, that is, a storage device that generates final data frame information for the higher level device, a flag indicating whether or not data of the user data segment is valid (Data Valid bit) It is preferable to delete the area and transmit it to the host apparatus as a payload composed of only user data.

In addition, as a transmission process of the status information to the host device 2, a storage device of the last stage that performs the data transmission process, that is, a storage device that generates final data frame information for the host device, is received from the preceding device. The generation of status information for the host apparatus 2 is realized by generating and transmitting new status information from one status information and its own status information.

(2-3) Write processing (data update processing in one storage device)

FIG. 30 is a diagram showing an example of data transfer processing in the case where the access request from the host device 2 is a write data request and the data update processing of one storage device and redundant data update processing of one storage device are performed. In FIG. 30, the storage device 1-1 performs the data update process (Data10 → Data1N), and the storage device 1-N performs the redundant data update process (DataP0 → DataPN).

As described above, each of the storage devices 1-1 to N independently processes the access request based on the access request information stored in the reception command storage 26 and the cooperative control information 29. It is determined whether or not to perform redundant data update processing according to the access request, and the disk device which judges that it does not process is completed to complete the processing required for the write data request by deleting the access request.

In the storage device 1-1 which has determined that the data update process is performed by itself, the storage device 1-1 is provided with respect to the data to be updated (old data: Data10) stored in the disk 11-1 through the disk control unit 14-1. The disc read process starts.

Further, data (new data: Data1N) received from the host device 2 through the interface control unit 13-1 and update information (Data1N ') based on, for example, an exclusive OR operation result from the old data Data10. ) Is generated by the update information generation unit 16-1.

The Data Valid bit in which the scene data Data1N received from the host device 2 is replaced with the update information Data1N ', and " valid " is set to a flag corresponding to a segment storing the update information. The data frame information transmission process with the area added is performed.

In the storage device 1-1, the data update process is realized by storing the scene data Data1N in the disk unit 11-1 through the disk control unit 14-1.

In the storage device 1-1, after generating and transmitting the status information according to the above-described processing, the series of write data request processings are completed by deleting the access request.

In addition, in the storage device 1-N that determines that the redundant data update processing is performed, redundant data of the update target stored in the disk unit 11-N through the disk control unit 14-N (old redundant data) : Disc read processing for DataP0) is started.

In addition, the update information generation unit receives, from the update information Data1N 'received from the storage device 1-1 and the old redundancy data DataP0, for example, new redundancy data DataPN based on an exclusive OR operation result. The redundancy data update process is realized by generating (16-N) and storing the redundant data (DataPN) in the disk unit 11-N via the disk control unit 14-N.

In addition, the storage device 1-N generates new status information based on the status information according to the above-described processing and the status information received from the preceding device, and performs a transmission process on the host device 2, and then performs the above-described processing. By deleting the access request, a series of read data request processings are completed.

In the redundant data update process, it is detected whether or not a flag corresponding to the Data Valid bit area is set to "valid", and the redundant data update process is performed only when it is set to "valid". If the "valid" setting remains, the received data frame information is retransmitted to the device at a later stage, so that, for example, the storage device as a data update processing target is located at the rear of the storage device performing redundancy data update. In this case, the data update process / redundant data update process can be realized.

(2-4) Write processing (data update processing in multiple storage devices)

FIG. 31 shows an example of the data transfer process when the access request from the host device 2 is a write data request and performs data update processing of a plurality of storage devices and redundant data update processing of one storage device. In FIG. 31, the storage device 1-1 / 2 performs data update processing (Data10 / 20-> Data1N / 2N), and the storage device 1-N performs redundant data update processing (DataP0-> DataPN). I do it.

As described above, each of the storage devices 1-1 to N independently processes the access request based on the access request information stored in the reception command storage 26 and the cooperative control information 29. The disk device which judges whether the redundancy data update process according to the access request is to be performed or not, and judges that it does not process, deletes the access request, thereby completing the process that takes the write data request.

In the storage device 1-1 which has determined that the data update process is performed by itself, the storage device 1-1 is used to update data (old data: Data10) stored in the disk unit 11-1 through the disk control unit 14-1. The disc read process is started.

In addition, among the data (new data: Data1N / 2N) received from the host device 2 through the interface control unit 13-1, from the own new data Data1N and the old data Data10, for example, The update information generation unit 16-1 generates update information Data1N 'based on the result of the exclusive OR operation.

Further, among scene data Data1N / 2N received from the host device 2, scene data Data1N for itself is replaced with the update information Data1N ', and corresponding to a segment for storing the update information. The data frame information transmission process in which the Data Valid bit area in which "valid" is set for a flag and "valid" is set for a flag corresponding to another user data segment is performed.

In addition, the storage device 1-1 performs data update processing, status transmission processing, and the like as described above, and completes a series of write data request processing.

In addition, in the storage device 1-2 which has determined that the data update process is performed by itself, the update target data stored in the disk unit 11-2 via the disk control unit 14-2 (old data: Data20). Disc read processing is started.

Further, among the data (update information: Data1N ', scene data: Data2N) received from the front-end storage device 1-1 via the interface control unit 13-2, scene data (2) for itself and the phrase The update information generation unit 16-2 generates update information Data2N 'based on, for example, an exclusive OR operation result from the data Data2O.

Furthermore, of the data (Data1N ', Data2N) received from the storage device 1-1 in the previous stage, a segment for replacing the scene data Data2N for itself with the update information (Data2N') and storing the update information. The data frame information transmission process with the flag corresponding to " valid " is performed.

Further, as described above, the data update processing and the generation and transmission processing of the new status information from the status information and the own status information received from the preceding apparatus are executed, and the series of write data request processing is completed by deleting the access request. do.

In addition, in the storage device 1-N that has determined that the redundant data update process is performed, redundant data of the update target stored in the disk unit 11-N via the disk control unit 14-N (formerly Disc read processing for redundant data: DataP0) is started.

The renewal length data DataPN based on, for example, an exclusive OR operation result is updated from the update information Data1N '/ 2N' received from the storage device 1-2 and the old length data DataPO. The redundant data update process is realized by generating the redundant redundant data DataPN in the information generating section 16-N and storing the redundant redundant data DataPN in the disk section 11-N via the disk control section 14-N.

In addition, the storage device 1-N generates new status information based on the status information according to the above-described processing and the status information received from the storage device 1-1 / 2 which has performed the above-described data update processing, After performing the transmission process on the apparatus 2, the series of write data request processes are completed by deleting the access request.

The redundant data update processing is performed based on the setting value of the flag corresponding to the Data Valid bit area as described above.

Next, the above-described read data request processing and write data request processing in the storage device 1 constituting the storage device subsystem 4 will be described.

32 is a diagram illustrating an example of a processing flow in the storage device 1 for a read data request from the host device 2.

The storage device 1 determines whether or not the access request is handled by itself based on the cooperative control information 29 and the access request information stored in the received command storage 26 (data transmission necessity or unnecessary). Discrimination processing is performed (step 100).

The storage device 1, which determines that it has not processed, performs the access request deletion (step 106) described later, and completes the processing required for the read data request.

The storage device 1, which has determined that it is processing, starts the disk read process from the disk unit 11 to the transmission / reception data storage unit 27 of the buffer unit 12 via the disk control unit 14 (step 101) Further, on the basis of the cooperative control information 29 and the access request information, a determination process is performed as to whether or not it is permitted to start read data transmission (read data transmission start / absence detection) (step 102). ).

The storage device 1, which has been permitted to start the transmission of read data, stores its own read data in the user data segment 0, dummy data in another user data segment, and a flag corresponding to the user data segment 0. The data frame information is set to " valid " and " invalid " to a flag corresponding to another user data segment, and the transmission process is performed for the device of the next stage (step 103).

Here, when there is no storage device which cooperatively processes the access request from the host device 2 other than the storage device 1 which has been permitted to start the transmission of read data, the storage device 1 Does not set / add the flag information (Data Valid bit area), generates data frame information composed of read data only, and performs transmission processing to the host apparatus 2.

On the other hand, the memory device 1 which is forbidden to start the transmission of read data by itself waits for the reception of the read data frame from the storage device of the preceding stage (step 107), and the reception of the read data frame is initiated ( Step 108), update processing of the received lead data frame (saves its own lead data in the user data segment to which its lead data should be stored, and also sets "valid" to a flag corresponding to that user data segment). Then, the transmission process of the updated read data frame is performed (step 109).

Here, in the storage device 1 in which the start of the transmission of the read data is prohibited, the storage device 1 which stores its read data in the last segment of the user data segment has the flag information (Data Valid bit area). ), Data frame information composed only of read data is generated, and the transmission process to the host apparatus 2 is performed.

In addition, the storage device 1 that has determined that the processing is to process the completion report transmission start of the access request processing based on the cooperative control information 29 and the access request information stored in the received command storage unit 26. A discrimination process is performed as to whether or not it is permitted (completion report transmission start or not detection) (step 104).

The storage device 1, which has been granted the start of the transmission reported by itself, generates the status information according to the read data process, performs a transmission process for the device of the next stage (step 105), and also transmits the status information. Then, by deleting the access request (step 106), a series of read data request processings are completed.

On the other hand, the memory device 1 forbidding the start of the transmission reported by the self completion is waited for reception of the status information from the storage device of the previous stage (step 110), and the reception of the status information is triggered (step 111). ), The process of updating the received status information (reflection of its own read data processing result) is performed, and the process of transmitting the updated status information is executed (step 112), and then the access request is deleted (step 106). This completes the read data request processing.

In the above-described embodiment, whether or not to store its own lead data in the user data segment 0 is taken as an example of the condition for detecting lead data transmission start or not, and its own lead data is stored in the user data segment 0. In the storage device described above, the data frame information added with the flag information (Data Valid bit area) has been described as being generated, but the present invention is not limited thereto. Similarly, in the storage device 1 storing its lead data in the last segment of the user data segment, the flag information (Data Valid bit area) is deleted and data frame information composed of lead data is generated. Although it demonstrated, it is not only this, but it is possible to determine the said storage device by arbitrary algorithms.

33 is a diagram illustrating an example of a processing flow in the storage device 1 for a write data request from the host device 2.

The storage device 1 determines whether or not the access request is to be processed (data update) based on the cooperative control information 29 and the access request information stored in the received command storage 26. A determination process of whether or not to perform redundant data update processing in accordance with the access request is performed (step 200).

The storage device 1, which determines that it does not process, performs the access request deletion (step 209) described later, and completes the process that takes the write data request.

The storage device 1, which has determined that the data update process is performed by itself, starts the disc read process with respect to the data (old data) to be updated stored in the disc unit 11 via the disc control unit 14 ( Step 201).

In addition, waiting for reception of the write data frame information from the preceding device (step 202), and upon receiving the write data frame information (step 203), the write data (new data) constituting the received write data frame information The update information generation unit 16 generates update information from the scene data to be stored by itself and the old data by an exclusive OR operation or the like (step 204).

Further, among the write data (scene data collection) constituting the received write data frame information, the scene data for itself is replaced with the update information, and a flag corresponding to the user data segment storing the update information is "valid." The transmission process of the write data frame information set to &quot; &quot; is executed (step 205).

Here, when the received write data frame information does not add the flag information (Data Valid bit area), the storage device 1 applies "valid" to a flag corresponding to a user data segment storing its update information. Then, the write data frame information to which the flag information which set "invalid" to the flag corresponding to another user data segment is added is generated, and the write data frame information is transmitted.

In addition, write data update processing is performed by storing (disk writing) the scene data to be stored by the self in the disk unit 11 via the disk control unit 14 (step 206).

In addition, the storage device 1 that has determined that the redundancy data update process is to perform the disk read process for the redundant data (old redundancy data) to be updated stored in the disk unit 11 via the disk control unit 14. Start (step 210).

In addition, it waits for reception of write data frame information from the preceding device (step 211), and all the flag information (Data Valid bit area) corresponding to the user data segment constituting the write data frame information is "valid". On the basis of detection of what has been set (step 212), the update information generation unit 16 stores update information (aggregate) constituting the write data frame information and credit length data by an exclusive OR operation from the old book data. In step 213. In addition, although all the flag information was used as a condition here, it is good also as flag information which satisfy | fills a predetermined condition. By storing the redundancy data in the disk unit 11 via the disk control unit 14, redundancy data update processing is realized (step 214).

If all the flag information corresponding to the user data segment constituting the write data frame information is not set to "valid", it is assumed that the received write data frame is retransmitted.

This makes it possible to realize the generation / transmission process of the update information by the data update target storage device even when the data update target storage device is located at the rear end of the redundant data update target storage device.

Further, by the processing, the storage device that has received the write data frame information again has a flag corresponding to the user data segment to be stored, for example, based on the flag information constituting the write data frame information. Resend processing is performed on condition that it is set to "valid."

In addition, the storage device 1 that has determined that it performs data update processing or redundant data update processing is based on the cooperative control information 29 and the access request information stored in the received command storage unit 26. A determination process is performed as to whether or not the completion report transmission start of the access request processing is permitted (completion report transmission start or not detection) (step 207).

The storage device 1, which has been permitted to start the transmission reported by itself, generates the status information according to the write data processing, performs the transmission processing for the device of the next stage (step 208), and also transmits the status information. Then, by deleting the access request (step 209), a series of write data request processing is completed.

On the other hand, the memory device 1 forbidding the start of the transmission reported by the self completion is awaited to receive the status information from the storage device of the previous stage (step 215), and the reception of the status information is initiated (step 216). ), The process of updating the received status information (reflecting the result of its own write data processing) is performed, and the process of transmitting the updated status information is executed (step 217), and then the access request is deleted (step 209). This completes the write data request process.

In the above-described embodiment, since the redundant data update processing is performed after the data update processing is performed in all the storage devices targeted for the data update process, the redundant data update destination is the final status. Although information is generated / transmitted, it is not limited to this. Similarly, the method of generating the status information by the individual storage device has been described as an example of the method of generating the final status information. However, not only this but the specific storage device constituting the storage device subsystem 4 is described. It is also possible to manage the generation transmission of the status information.

[Fifth Embodiment] (Degenerate Operation)

EMBODIMENT OF THE INVENTION Hereinafter, 5th Embodiment which concerns on this invention is described using FIGS. 34-17.

In the fifth embodiment, the storage device of the fourth embodiment adds and maintains operation state information of the storage device other than the cooperative processing, so that the host device even if a failure occurs in the storage device constituting the storage subsystem. To implement access request processing for the storage subsystem.

Fig. 34 is information for configuring redundant configuration management information of the cooperative control information 29 used in the description of the above-described embodiment, which is " 5 " for RAID level information and " 3 / " for data disk number / redundant data disk number information. 1 &quot; is set, the operation mode information and the failure memory device identification information (the failure memory device ID) are held, and the degenerate mode due to the storage device failure of ID_2 is set.

In addition, since the storage device in this embodiment has the same configuration as that described above, the drawings are omitted.

In addition, when the data restoration process mentioned later is unnecessary, since it can implement by the control similar to embodiment mentioned above, description is abbreviate | omitted.

Hereinafter, a case where a failure occurs in the storage device 1-2 constituting the storage device subsystem 4 is described as an example. The access request from the host device 2 to the storage device system 4 is described above. The operation of processing in cooperation with other storage devices except the storage device 1-2 will be described.

(1) Read processing for obstacle storage device (data recovery processing is realized by multiple storage devices)

FIG. 35 shows data transfer when a read data request from the host device 2 for data stored in the storage device 1-2 in a fault state is processed by a storage device other than the storage device 1-2. FIG. It is a figure which shows an example of a process.

Storage devices other than the storage device 1-2 stored the access request information in the received command storage unit 26 process the access request based on the access request information and the cooperative control information 29. In this case, it is detected that the restoration process of the data stored in the storage device 1-2 in the fault state is necessary, and it is determined whether or not the data restoration process is performed by itself.

The storage device judged that the restoration process of the data stored in the storage device in the disturbed state is necessary, if necessary, the data necessary for the data restoration process stored in the disc unit 11 via the disc control unit 14 (Fig. In the meantime, the disc read process for the Data1, Data3, and Data4 is started.

In addition, in the following description, based on the access request information and the cooperative control information 29, the storage device located at the foremost (fastest receiving information from the host device) the data according to the data restoration process. Although the case where the storage device which starts transmission of the data transmission and which is located at the last end (receiving information from the host device last) performs the above data restoration process is described as an example, it is not limited to this.

In the storage device 1-1, which has been permitted to start data transmission in accordance with the data restoration process, for example, it stores its own data in the user data segment 0, and also flags corresponding to the user data segment 0. By performing the transmission process of the data frame information in which " valid " is set to " invalid " in a flag corresponding to another user data segment, and deleting the access request after generating and sending status information according to the processing. The series of read data request processes is completed.

In addition, in the storage device 1-3, which is forbidden to start data transmission according to the data restoration process, the user data whose own data is stored in the data frame information received from the storage device 1-1. A segment 1 is added, and a data frame information generation / transmission process with " valid " set to a flag corresponding to the user data segment 1 is performed, and the status information according to the process and the status received from the preceding device. After generating and transmitting new status information based on the information, a series of read data request processes are completed by deleting the access request.

Further, in the storage device 1-4 which determines that the data restoration process is performed by itself, the data Data1 and Data3 received from the storage device 1-1 / 3 and the data stored in the storage device 1-4 are stored in the data restoration process. From the required data (Data4), by the exclusive OR operation in the update information generation unit 16-4 or the like, a restoration process of the data Data2 stored in the storage device 1-2 in the fault state is performed. Using the reconstructed data Data2, the host device 2 transmits lead data frame information.

In addition, a flag indicating whether or not the data of the user data segment is valid in the storage device of the last stage that performs the data restoration process, that is, the storage device that generates data frame information of the last type for the higher level device (Data Valid bit) The area is preferably deleted and transmitted to the host device as a payload composed only of user data.

In addition, as a transmission process of the status information to the host device 2, the storage device of the last stage which performs the said data transmission process, ie, the memory device which produces | generates the last-type data frame information about the host device, is a front end. The generation of status information for the host apparatus 2 is realized by generating and transmitting new status information from the status information received from the device and the own status information.

In addition, in the above-described embodiment, the read data request from the host device 2 is processed in a storage device other than the storage device 1-2 only with respect to the data stored in the storage device 1-2 in the fault state. Although the case has been described as an example, not only this but, for example, in the case of a read data request for data stored in the memory device 1-1 / 2, the memory device 1-4 is characterized in that the memory device ( It is possible to generate / transmit the data frame information of the final type for the higher level device using the data Data1 received from 1-1) and the data Data2 restored by itself.

(2) Write processing for obstacle storage device (redundant data update processing in multiple storage devices)

FIG. 36 shows an example of data transfer processing when the write data request from the host device 2 to the storage device 1-2 in the fault state is processed by a storage device other than the storage device 1-2. It is a figure which shows.

A storage device other than the storage device 1-2 that has stored the access request information in the received command storage section 26, based on the access request information and the cooperative control information 29, performs a data update process by itself. It is determined whether or not to perform the redundant data update process according to the access request.

If it is determined that the data update process or redundant data update process is not performed, the storage device, if necessary, the data necessary for the redundant data update process stored in the disk unit 11 through the disk control unit 14 (in the drawing). Disc read processing for, Data1, Data3) is started.

In addition, in the following description, the storage device located at the foremost end (the earliest receiving information from the host device) starts the transmission of data according to the redundant data update process, and is located at the end (from the parent device). Although the storage device (receiving the information of) last performs the data restoration process as an example, the present invention is not limited thereto.

In the storage device 1-1, which has been permitted to start data transmission according to the redundant data update process, for example, its own data Data1 has been received from the host device 2 in the user data segment 0. The data in which scene data Data2 is stored in the user data segment 1, and " valid " is set to a flag corresponding to the user data segment 0/1 and " invalid " to a flag corresponding to another user data segment. A series of write data request processings are completed by performing the frame information transmission process and deleting the access request after generating and transmitting the status information according to the processing.

In addition, in the storage device 1-3, which is forbidden to start data transmission in response to the redundant data update process, it stores its own data Data3 in the data frame information received from the storage device 1-1. A user data segment 2 is added, and a data frame information generation / transmission process with " valid " set to a flag corresponding to the user data segment 2 is performed, and further from the status information and the preceding device according to the process. After generating and transmitting new status information based on the received status information, the series of write data request processing is completed by deleting the access request.

Further, in the storage device 1-4 which has determined that the redundant data update process is performed by itself, from the data Data1, Data2, Data3 finally received from the storage device 1-3, the update information generation unit The new redundant data DataP is generated by an exclusive-OR operation in (16-4), and the generated redundant data DataP is transferred to the disk unit 11-4 through the disk control unit 14-4. By storing in the message, redundant data update processing is realized.

In addition, the storage device 1-4 generates new status information based on the status information according to the above-described processing and the status information received from the preceding device, and performs a transmission process on the host device 2, and then By deleting the access request, a series of write data request processing is completed.

In the redundant data update process, it is detected whether or not a flag corresponding to the Data Valid bit area is set to "valid", and the redundant data update process is performed only when it is set to "valid". If the "valid" setting remains, the received data frame information is retransmitted to the device at a later stage, so that, for example, the storage device as a data update processing target is located at the rear of the storage device performing redundancy data update. In this case, the redundant data update process can be realized.

In addition, in the above-described embodiment, the write data request from the host device 2 is processed in a storage device other than the storage device 1-2 only with respect to the data stored in the storage device 1-2 in the fault state. Although the case has been described as an example, not only this but for example, in the case of a write data request for data stored in the storage device 1-1 / 2, the above-described disk read in the storage device 1-1 is described. Of the scene data (Data1, Data2) received from the host device (2) without performing the process, the scene data (Data1) for itself is sent to the disk unit (11-1) via the disk control unit (14-1). Storing data in the user data segment 0/1 and storing the scene data Data1 and Data2 received from the host device 2, and also corresponding to the user data segment 0/1. "Valid" on the flag, other A series of write data request processes are performed by performing the transmission processing of the data frame information in which the flag corresponding to the user data segment is set to "invalid", and after generating and transmitting the status information according to the processing, and deleting the access request. It is possible to complete.

In the cooperative control information 29, it is preferable that the cooperative control information 29 can be set from the host apparatus, and it is preferable that the operation mode information or the like can be set at its own discretion.

Sixth Embodiment (Multi Pass Control)

Hereinafter, a sixth embodiment according to the present invention will be described with reference to FIGS. 37 and 19.

In the sixth embodiment, the storage device of the above-described embodiment further includes a plurality of interface control units, and transfers the access request information from the host device and the user data transferred between the storage devices constituting the storage subsystem. The processing is realized by using another interface (transmission path).

FIG. 37 is a diagram illustrating an example of an information processing system composed of a storage device subsystem 4 composed of a plurality of (N, N> 1) storage devices 1 and an upper level device 2.

In FIG. 37, the individual storage devices 1 constituting the storage device subsystem 4 are provided with a plurality of interface control units A / B (13-a / b), and the interface control unit A (13-a). Similarly to the above-described embodiment, the host device 2 is directly connected to the host device 2 through the circuit 3a of the circulation type. At the same time, the individual storage devices 1 constituting the storage device subsystem 4 are connected via the circulation type interface 3-b using the interface control unit B 13-b.

In addition, as an example of the configuration of the information processing system, only the individual storage devices 1 constituting the storage device subsystem 4 are used using the interface control unit B 13-b, and the cyclic interface 3-b. Although it is set as the structure connected through, it is not limited to this as mentioned later.

Hereinafter, in a state where a failure occurs in the storage device 1-2 constituting the storage device subsystem 4, a read from the host device 2 to the data stored in the storage device 1-2 in the failed state is performed. The operation will be described taking as an example the case where the data request is cooperatively processed by another storage device except for the storage device 1-2.

38 shows data transfer when a read data request from the host device 2 for data stored in the storage device 1-2 in a fault state is processed by a storage device other than the storage device 1-2. It is a figure which shows an example of a process.

A storage device other than the storage device 1-2 stored the access request information in the received command storage unit 26 via the interface control unit A 13-a, as in the above-described embodiment, the access request information. And on the basis of the cooperative control information 29, in processing the access request, it is detected that the restoration process of the data stored in the storage device 1-2 in the disturbed state is necessary, and the data restoration process is performed by itself. Determine whether or not to execute.

The storage device judged that the restoration process of the data stored in the storage device in the disturbed state is necessary, if necessary, the data necessary for the data restoration process stored in the disc unit 11 via the disc control unit 14 (in the drawing, Disc read processing for Data1, Data3, Data4) is started.

In the storage device 1-1, which has been permitted to start data transmission in accordance with the data restoration process, for example, it stores its own data Data1 in the user data segment 0, and further stores in the user data segment 0. The transmission control of the data frame information in which "valid" is set to a corresponding flag and "valid" to a flag corresponding to another user data segment is used for an interface control unit B different from the interface control unit A which has received the access request information. And after generating and transmitting the status information according to the processing, a series of read data request processings are completed by deleting the access request.

In addition, in the storage device 1-3, which is forbidden to start data transmission according to the data restoration process, the user data whose own data is stored in the data frame information received from the storage device 1-1. A generation / transmission process of data frame information in which segment 1 is added and " valid " is set to a flag corresponding to the user data segment 1, and also based on the status information according to the process and the status information received from the preceding apparatus. After performing the generation and transmission processing of the new status information to be performed using the interface control unit B different from the interface control unit A that has received the access request information, similarly to the storage device 1-1, a series of steps are deleted by deleting the access request. The read data request processing is completed.

In addition, in the storage device 1-4 which has determined that the data restoration process is performed by itself, the data (Data1, Data3) received from the storage device 1-1 / 3 through the interface control unit B 13-4b. ), And from the data (Data4) stored and necessary for the data restoration process, to the storage device (1-2) in the disturbed state by an exclusive OR operation in the update information generation unit (16-4). A process of restoring the stored data Data2 is performed, and the process of transmitting the lead data frame information to the host apparatus 2 is performed using the interface controller A 13-4a. will be.

In addition, although the detailed description is omitted, similarly in the write data processing, the data transfer processing that cooperates between the storage devices forming the storage subsystem is performed by using an interface different from the interface receiving the access request information from the host device. It is possible to realize.

In addition, in the above description, the case where cooperative data transfer process is performed between the storage devices which comprise the storage subsystem using an interface different from the interface which received the access request information from the host device is used as an example. It does, but is not limited to this. For example, it is also possible to select an interface for performing data transfer processing between the storage apparatuses performing the cooperative processing on the basis of the operating state of the interface at the time of frame transmission, and on the interface for performing the data transfer processing. Even if the host device 2 is connected, there is no problem.

Each embodiment described above has the following effects.

According to the first embodiment of the present invention, in a storage system composed of a plurality of storage devices, the individual storage devices constituting the storage system request access to the storage device system from a higher level device. Means for sharing in the storage device, means for determining whether or not to process the shared access request by the storage device, and also for the storage device that determines that at least the write data transmitted from the host device to the storage system should be processed. By providing the means for transferring, a control device for controlling individual storage devices is unnecessary, and a storage device system in which the use efficiency of the storage device is expected to be improved can be realized.

According to the second embodiment of the present invention, in a storage system composed of a plurality of storage devices, at least one storage device constituting the storage system receives an access request to the storage device from a higher level device, The storage device further includes means for determining which storage device should process the access request and, if necessary, transferring the storage device from the higher-level device to the storage device that should process the access request to the storage device system. It is possible to realize a storage system in which a control device for controlling individual storage devices is not necessary and the improvement in utilization efficiency of the storage device is expected.

According to a third embodiment of the present invention, in a storage system composed of a plurality of storage devices, the means for allowing individual storage devices constituting the storage system to share write data requests to the storage device from at least an upper level device. And means for synchronizing the transfer process of write data and the transfer process of the status information for the write data request with respect to the shared write data request in synchronization with each other. A control device for controlling the storage device is unnecessary, and a storage system for reducing the interface load can be realized.

According to the present invention, in a storage system including RAID1 composed of a plurality of storage devices, an access request from a higher level device can be cooperatively processed between a plurality of storage devices that constitute the storage system.

BRIEF DESCRIPTION OF THE DRAWINGS The figure which shows one structural example of the information processing system of this invention.

2 is a diagram showing an example of the configuration of a memory device according to the first embodiment of the present invention.

3 is a diagram illustrating an example of a read processing flow according to the first embodiment of the present invention.

4 is a diagram showing an example of information transmission in a read process according to the first embodiment of the present invention.

5 is a diagram illustrating an example of a light processing flow according to the first embodiment of the present invention.

Fig. 6 is a diagram showing an example of information transmission in the write process according to the first embodiment of the present invention.

Fig. 7 is a diagram showing an example of configuration of unupdated area management information according to the first embodiment of the present invention.

Fig. 8 is a diagram showing an example of information transmission in the update process of the unupdated area according to the first embodiment of the present invention.

9 is a diagram showing one configuration example of a memory device according to the second embodiment of the present invention.

10 is a diagram illustrating an example of a read processing flow according to the second embodiment of the present invention.

Fig. 11 is a diagram showing an example of information transmission in read processing according to the second embodiment of the present invention.

12 is a diagram showing an example of a light processing flow according to the second embodiment of the present invention.

Fig. 13 is a diagram showing an example of information transmission in the update process of the unupdated area according to the second embodiment of the present invention.

14 is a diagram showing an example of a light processing flow according to the third embodiment of the present invention.

Fig. 15 is a diagram showing an example of information transmission in the write process according to the third embodiment of the present invention.

FIG. 16 is a diagram illustrating an example of configuration of cooperative control information of the present invention. FIG.

17 is a diagram illustrating an example of a configuration of access request information.

18 is a diagram illustrating an example of a configuration of frame header information.

Fig. 19 is a diagram showing one configuration example of the write processing management information of the present invention.

20 is a diagram illustrating an example of the configuration of an information processing system of the present invention.

21 is a diagram showing one configuration example of a memory device according to the fourth embodiment of the present invention.

Fig. 22 is a diagram showing a configuration and setting example of cooperative control information according to the fourth embodiment of the present invention.

Fig. 23 is a diagram showing one configuration example of frame information used in the present invention.

Fig. 24 is a diagram showing one configuration example of frame information constituting frame information used in the present invention.

Fig. 25 is a diagram showing one configuration example of access request information constituting frame information used in the present invention.

Fig. 26 is a diagram showing an example of a transmission processing flow of access request information in the present invention.

Fig. 27 is a diagram showing an example of read data transfer processing (single) according to the fourth embodiment of the present invention.

Fig. 28 is a diagram showing an example of read data transfer processing (plural) according to the fourth embodiment of the present invention.

Fig. 29 is a diagram showing an example of configuration of data frame information in the present invention.

Fig. 30 is a diagram showing an example of write data transfer processing (single) according to the fourth embodiment of the present invention.

Fig. 31 is a diagram showing an example of write data transfer processing (plural) according to the fourth embodiment of the present invention.

32 is a diagram showing an example of a read processing flow according to the fourth embodiment of the present invention.

33 is a diagram showing an example of a light processing flow according to the fourth embodiment of the present invention.

Fig. 34 is a diagram showing a configuration and setting example of cooperative control information according to the fifth embodiment of the present invention.

Fig. 35 is a diagram showing an example of read data transfer processing according to the fifth embodiment of the present invention.

36 is a diagram showing an example of write data transfer processing according to the fifth embodiment of the present invention;

37 is a diagram showing an example of the configuration of an information processing system according to a sixth embodiment of the present invention.

38 is a diagram showing an example of read data transfer processing according to the sixth embodiment of the present invention.

<Explanation of symbols for the main parts of the drawings>

1: storage device

2: parent device

3: Interface (SAN: Storage Area Network)

4: storage system

10: main control unit

11 disc portion (memory medium)

12: buffer part

13 interface control

14: disc controller

15: buffer control unit

20: receiver

21: transmitter

22: receiving frame processing unit

23: transmission frame generation unit

24: retransmission judgment unit

25: output selector

26: Receive command storage unit

27: transmission and reception data storage unit

28: cooperative control information

29: command transmission control unit

30: Unupdated Zone Management Information

Claims (24)

In a storage system including a plurality of storage devices connected to the host device, Each of the plurality of storage devices, Means for receiving a request for information processing of the storage system in the host device from the host device or other storage devices included in the plurality of storage devices; Means for determining whether the received request is a request to be processed in a corresponding storage device (hereinafter referred to as a 'first storage device') that has received the request; Means for transferring the received request to a second other storage device (hereinafter referred to as a 'second storage device') included in the plurality of storage devices when the request is not processed in the first storage device. and, Means for executing information processing indicated by the received request, in the case where the received request is a request to be executed by the first storage device. The method of claim 1, The means for executing the information processing performs the information processing when it is determined that the received request should be executed based on the cooperative control information indicating the request to be executed by the first storage device and the received request. Memory system running. The method of claim 2, The request includes first identification information indicating a storage device to which the information processing is to be executed, the cooperative control information includes second identification information for identifying the first storage device, And the means for executing the information processing executes the information processing when the first identification information and the second identification information correspond. The method of claim 1, The second memory device Means for receiving the transmitted request, and Means for executing information processing indicated by the transmitted request when the transferred request is a request to be executed in the second storage device. Storage system comprising a. The method of claim 4, wherein In the first storage device, information indicating the first storage device is added to a request transmitted by the transferable means. And the second storage device further comprises means for suppressing transmission of the transferred request again based on the information indicating the added first storage device. The method of claim 1, The transferable means is further configured to transmit the received request when it is determined that it is not necessary to execute the received request based on the cooperative control information indicating the request to be executed by the first storage device and the received request. 2 A storage system that transfers to storage. The method of claim 1, The transferable means is based on the cooperative control information indicating a request to be executed by the first storage device and the received request when it is determined that the second storage device should execute the received request. A storage system for transmitting a request to said second storage device. The method of claim 7, wherein The request includes first identification information indicating a storage device to which the information processing is to be executed, the cooperative control information includes second identification information for identifying the first storage device, And means for executing the information processing executes the information processing when the first identification information and the second identification information correspond. In a storage system including a plurality of storage devices connected to the host device, Each of the plurality of storage devices, A receiving unit, connected with other storage devices included in the host device or the plurality of storage devices, for receiving a request for information processing from the storage device to the storage device system; A transmitter connected to a second memory device (hereinafter referred to as a second memory device) included in the receiver and the plurality of memory devices, and transmitting the received request to the second memory device; and If the received request is to be executed by the corresponding storage device (hereinafter referred to as 'first storage device'), includes a processor that executes the information processing indicated by the received request; And the processor determines whether the received request should be executed in the first storage device. The method of claim 9, And the processor executes the information processing when it is determined that the received request is to be executed based on the cooperative control information indicating the request to be executed by the first storage device and the received request. The method of claim 10, The request includes first identification information indicating a storage device to which the information processing is to be executed, the cooperative control information includes second identification information for identifying the first storage device, And the processor executes the information processing when the first identification information and the second identification information correspond. The method of claim 9, The second memory device, A second receiver connected to said transmitter, for receiving said transmitted request, and And a second processor connected to the second receiving unit and executing information processing indicated by the transmitted request when the transmitted request is a request to be executed in the second storage device. The method of claim 12, In the first storage device, information indicating the first storage device is added to a request transmitted by the transmission unit. And the second processor inhibits transmission of the transmitted request again based on the information indicating the added first storage device. The method of claim 9, The transmitting unit transmits the received request to the second request when it is determined that it is not necessary to execute the received request based on the cooperative control information indicating a request to be executed by the first storage device and the received request. Storage system to transfer to storage. The method of claim 9, The transmitting unit receives the request if it is determined that the second storage device should execute the received request based on the cooperative control information indicating the request that the first storage device should execute and the received request. A storage system for transmitting the data to the second storage device. The method of claim 15, The request includes first identification information indicating a storage device to which the information processing is to be executed, the cooperative control information includes second identification information for identifying the first storage device, And the processor executes the information processing when the first identification information and the second identification information correspond. In the storage control method in each storage device included in the storage system including a plurality of storage devices connected to the host device, Executed in each of the above storage devices, Receiving a request for the information processing of the storage system in the host device from another storage device of the host device or the plurality of storage devices, Determining whether the received request is a request to be processed in a corresponding storage device (hereinafter referred to as a 'first memory device') that has received the request, If the request is not processed in the first storage device, transmitting the received request to a second other storage device (hereinafter referred to as a 'second storage device') included in the plurality of storage devices, and And executing the information processing indicated by the received request if the received request is a request that the first storage device should execute. The method of claim 17, The step of executing the information processing is based on the cooperative control information indicating the request to be executed by the first storage device and the received request, when it is determined that the received request should be executed. Memory control method. The method of claim 17, The request includes first identification information indicating a storage device to which the information processing is to be executed, the cooperative control information includes second identification information for identifying the first storage device, The step of executing the information processing is a storage control method that executes the information processing when the first identification information and the second identification information correspond. The method of claim 17, Executed in the second storage device, Receiving the transmitted request, and And executing the information processing indicated by the transmitted request, if the transmitted request is a request to be executed in the second storage device. The method of claim 20, In the first storage device, in the transferable step, information indicating the first storage device is added to a request to be transmitted, Inhibiting transmission of the transferred request again based on the information indicating the added first memory device executed in the second memory device. The method of claim 17, The step of transmitting the request is based on the cooperative control information indicating a request to be executed by the first storage device and the received request when it is determined that the received request is not to be executed. 2 Memory control method to transfer to memory. The method of claim 17, The transmittable step is based on the cooperative control information indicating a request to be executed by the first storage device and the received request when it is determined that the second storage device should execute the received request. A storage control method for transmitting a request to said second storage device. The method of claim 23, wherein The request includes first identification information indicating a storage device to which the information processing is to be executed, the cooperative control information includes second identification information for identifying the first storage device, The step of executing the information processing is a storage control method that executes the information processing when the first identification information and the second identification information correspond.