JP2014164431A - Redundant system, redundant method, storage device, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a redundant system capable of realizing power saving with maintained redundancy even when a failure occurs in a path, a redundant method, a storage device, and a program.SOLUTION: The redundant system comprises a disk array device 1, and a business hosts 5 and 6 connected to the disk array device 1 through a plurality of first paths. Also, the redundant system comprises port control means 12, and controller control means 14. The port control means 12 acquires first path connection information which shows whether the plurality of first paths can communicate. The controller control means 14 stops operation of a controller so as to maintain the connection between the disk array device 1 and the business hosts 5 and 6 using at least two first paths which can communicate on the basis of the first path connection information.

Description

  The present invention relates to a redundancy system, a redundancy method, a storage device, and a program.

  2. Description of the Related Art Conventionally, storage devices that make up a disk pool from a set of magnetic disks such as HDD (Hard Disk Drive) such as a disk array represented by RAID (Redundant Arrays of Inexpensive Disks) are known. Further, for example, as in Patent Document 1, there is a system that uses a logical disk obtained by logically dividing an area on a disk pool of a storage device from a business host connected by a network (for example, SAN (Storage Area Network)). To do.

  Generally, a plurality of controllers that control RAID are installed in a disk array device, and a redundant configuration is adopted in which a business host connects to a plurality of controllers on the disk array device through redundant paths. It has been.

  On the other hand, the number of systems for which the upper limit of power consumption is determined has increased in recent years. In the case of a server, there is an example in which an upper limit is provided for each rack, and there is a demand for reducing the power consumption of the disk array device. For example, Patent Document 2 discloses a technique for stopping a processor array and a hard disk in consideration of redundancy. Patent Document 3 discloses a technique for reducing the number of CPU clocks based on the busy rate of a CPU (Central Processing Unit).

JP 2001-216204 A JP 2012-181656 A JP 2008-250945 A

  However, if there is a path that is not normally connected between the disk array device and the business host (for example, a failure status), if a normal controller not related to the failure is stopped, redundancy cannot be maintained and May be affected.

  The present invention has been made to solve such a problem, and even when a failure occurs in a path, a redundancy system capable of realizing power saving while maintaining redundancy, An object is to provide a redundancy method, a storage device, and a program.

  The redundancy system according to the present invention includes a memory, a storage device that controls input / output of data to / from the memory and is connected to a plurality of first paths, and a plurality of the first devices. A first connection information acquisition means for acquiring first path connection information indicating whether or not a plurality of the first paths can communicate with a host device connected to the storage device via a path; A controller that stops the operation of the controller based on one path connection information so as to maintain the connection between the storage device and the host device using the first communicable first path. And a control means.

  A redundancy method according to the present invention includes a memory, a storage device that controls input / output of data to / from the memory and is connected to a plurality of first paths, and a plurality of the first devices. A system redundancy method comprising: a host device connected to the storage device via a path, wherein first path connection information indicating whether or not a plurality of the first paths can communicate is acquired And at least two or more communicable first paths based on the first path connection information, so that the connection between the storage device and the host device is maintained. And a step of stopping the operation.

  A storage device according to the present invention controls a memory, input / output of data to / from the memory, a plurality of controllers connected to a host device via a plurality of first paths, and a plurality of the first paths. First connection information acquisition means for acquiring first path connection information indicating whether or not communication is possible, and at least two or more first paths capable of communication based on the first path connection information. Controller control means for stopping the operation of the controller so as to maintain the connection between the used controller and the host device.

  A program according to the present invention controls a memory, a storage device that controls input / output of data to / from the memory, and is connected to a plurality of first paths, and a plurality of the first paths. A computer program for controlling a system including a host device connected to the storage device via the computer, and acquiring to the computer first path connection information indicating whether or not the plurality of first paths are communicable And the controller so as to maintain the connection between the storage device and the host device using at least two or more communicable first paths based on the first path connection information. The step of stopping the operation is executed.

  According to the present invention, it is possible to provide a redundancy system, a redundancy method, a storage device, and a program capable of realizing power saving while maintaining redundancy even when a failure occurs in a path. .

1 is a block diagram of a redundant system according to an embodiment. It is a figure which shows an example of the performance management table concerning embodiment. It is a figure which shows an example of the port state management table concerning embodiment. It is a figure which shows an example of the path | pass state management table concerning embodiment. It is a figure which shows an example of the host information management table concerning embodiment. It is a flowchart which shows operation | movement of the redundancy system concerning embodiment. It is a flowchart which shows operation | movement of the path control means concerning embodiment. 1 is a block diagram of a redundant system according to an embodiment.

  Embodiments of the present invention will be described below with reference to the drawings. The configuration of the redundant system according to this embodiment is shown in FIG. The redundant system includes a disk array device 1 (storage device), FC (fiber channel) switches 41 and 42 (relay devices), and business hosts 5 and 6 (host devices). The disk array device 1 is connected to the business hosts 5 and 6 through the FC switches 41 and 42 using a SAN network. The business host is a computer that provides services, performs arithmetic processing, etc. in response to a request from a client terminal (not shown).

  The disk array apparatus 1 includes a performance management unit 11, a port control unit 12, a path state monitoring unit 13, a controller control unit 14, a performance management table 15, a port state management table 16, and a path state management table 17. A host information management table 18, controllers 21 to 24, and a logical disk 3. The performance management unit 11, the port control unit 12, the path state monitoring unit 13, and the controller control unit 14 are realized, for example, by executing a processing program stored in advance by a CPU (Central Processing Unit) (not shown). . The performance management table 15, the port state management table 16, the path state management table 17, and the host information management table 18 are stored in a memory (not shown).

  The performance management means 11 (load information acquisition means) operates periodically and statistically measures the load information of the controllers 21 to 24. The load information is information indicating the load applied to each of the controllers 21 to 24. Further, the performance management unit 11 registers the measured value and the total value of the measured value in the performance management table 15.

  The operation of the performance management unit 11 will be described in detail. The performance management unit 11 periodically operates on the disk array device 1, acquires the load information of the controllers 21 to 24, and updates the performance management table 15. Specifically, the performance management unit 11 acquires the number of I / Os (IOPS (Input / Output Per Second)) per unit time as load information of each of the controllers 21 to 24 at an arbitrary time. As an IOPS measurement method, a counter of the number of I / O times is held, and a value obtained by dividing the difference value between the previous measurement and the current measurement by the interval time can be calculated as the IOPS at that time. The performance management unit 11 acquires IOPS measured in each of the controllers 21 to 24 and registers these values in the performance management table 15 as IOPS statistical values. Further, the performance management unit 11 calculates the sum of the IOPS statistical values of the controllers 21 to 24 and registers it in the performance management table 15 as the IOPS statistical value (total value) of the entire disk array device 1.

  FIG. 2 shows an example of the performance management table 15. The performance management table 15 stores I / O load statistical values and I / O load limit values of the controllers 21 to 24. The performance management table 15 stores I / O load statistical values and I / O load limit values for the entire disk array device 1. The average value of IOPS values for the past 24 hours is recorded in the statistical value of the I / O load in each of the controllers 21 to 24. The IOPS limit value records the performance limit IOPS value of each controller 21 to 24. Further, the sum of the values of all the controllers 21 to 24 is recorded in the statistical value and limit value (total value) of the entire disk array device 1.

  In the present embodiment, it is assumed that the IOPS limit values of the controllers 21 to 24 and the entire disk array device 1 are registered in advance by a vendor or a user. Note that the IOPS limit value may be updated manually or automatically in consideration of the possibility that the limit value fluctuates due to a change in the controller or the like. In the present embodiment, the IOPS value as described above is used as an example of the negative information of the controllers 21 to 24. However, the present invention is not limited to this. For example, it is conceivable to adopt an average over a longer period as a statistical value, or a performance guarantee limit value as a limit value. In addition to IOPS, an index for measuring the performance may be a CPU load (usage rate, etc.) of the controller, a method of using the cache memory usage rate, a method of comprehensively measuring them in combination.

  The port control means 12 (first connection information acquisition means) operates periodically, and path connection information (first path) indicating whether or not the paths connected to the ports of the controllers 21 to 24 are communicable. Connection information). The port control unit 12 registers the acquired path connection information in the port state management table 16.

  The operation of the port control means 12 will be described in detail. The port control unit 12 periodically operates on the disk array device 1 and registers the connection states (link states) of all the ports of the controllers 21 to 24 on the disk array device 1 in the port state management table 16. The port control unit 12 acquires connection information and failure information of all ports of each controller from the controllers 21 to 24. The port control means 12 sets “UP” when the port is in a link up state, “DOWN” when the port is in a link down state due to an unused port or a path failure, etc., as a link state of the corresponding port. Register in the management table 16. That is, the port control means 12 uses information indicating whether the path connected to the port is communicable (normal) or not communicable (abnormal) for all ports of each controller as port identification information. The data is stored in the port status management table 16 in association with each other.

  FIG. 3 shows an example of the port state management table 16. The port status management table 16 is a table for managing the connection status of each port of each controller 21 to 24. In this embodiment, among the ports of the controller 21, the port connected to the FC switch 41 is registered using the port name (identification information) of the controller 21-00, and the port connected to the FC switch 42 is the controller. It is registered using the port name 21-01. The connection status of each port is stored in association with the port name. As for the connection (link) state of each port, “UP” is registered as a link-up state when it is normally connected, and “DOWN” is registered as a link-down state when it is not connected or a failure has occurred. .

  The path status monitoring unit 13 (second connection information acquisition unit) operates periodically and acquires path connection information (second path connection information) of each path connected to all the business hosts 5 and 6. . That is, the path status monitoring unit 13 acquires path connection information of a path connecting the business hosts 5 and 6 and the FC switches 41 and 42. Further, the path status monitoring unit 13 registers the acquired path connection information in the path status management table 17.

  The operation of the path state monitoring unit 13 will be described in detail. The path status monitoring unit 13 periodically operates on the disk array device 1 and registers the connection status (link status) of all the paths of the business hosts 5 and 6 in the path status management table 17. The path status monitoring unit 13 confirms whether or not the FC protocol process login is performed for all the paths of the business hosts 5 and 6. Then, the path state monitoring means 13 sets “UP” as the link up state when the login is performed, and “DOWN” as the link down state when the login is not performed, and the path state management table 17 as the link state. Register with. That is, the path status monitoring unit 13 displays information indicating whether the path is communicable (normal) or not communicable (abnormal) for all paths connected to the business hosts 5 and 6. Are stored in the path status management table 17 in association with the identification information (host name).

  FIG. 4 shows an example of the path state management table 17. The path status management table 17 includes a host name for identifying the business hosts 5 and 6, a path name of a path that can recognize the logical disk 3 on the disk array device 1 from the business hosts 5 and 6, and connection of the paths. Information indicating the (link) state. In the host name, a name that can identify each of the business hosts 5 and 6 is stored. The HBA WWPN (World Wide Port Name) value is registered in the path name.

  In this embodiment, it is assumed that the host name and path (WWNP) are registered in advance by the user. As the connection status of each path, “UP” is registered as a link-up status when it is normally connected, and “DOWN” is registered as a link-down status when it is not connected or a failure has occurred. .

  The controller control means 14 operates periodically or when instructed to save the power of the controllers 21 to 24, and can stop the power supply without affecting the business performance based on the load state of each of the controllers 21 to 24. Determine the number of controllers. Further, the controller control unit 14 determines which controller can be maintained with the path redundancy based on the connection state of each port of the controllers 21 to 24 and the connection state of each path. And the controller control means 14 stops the power supply of the controller which can stop a power supply based on said determination result. Details of the operation of the controller control means 14 will be described later.

  The controllers 21 to 24 control data I / O on the disk array device 1. In the present embodiment, two paths are connected to one controller (see FIG. 1). Therefore, eight paths are connected as the entire disk array device 1. The logical disk 3 is a recording medium composed of memories such as a plurality of HDDs. Data is input / output to / from the logical disk 3 in accordance with the control of the controllers 21 to 24.

  The FC switches 41 and 42 are FC-connected to all host bus adapters (HBAs) on the network in order to form a redundant SAN network. The FC switch 41 is connected to each of the controllers 21 to 24 and also to each of the business hosts 5 and 6. Similarly, the FC switch 42 is connected to each of the controllers 21 to 24 and also to each of the business hosts 5 and 6. Note that. The path connecting the controllers 21 to 24 and the FC switches 41 and 42 is the first path, and the path connecting the FC switches 41 and 42 and the business hosts 5 and 6 is the second. Is the path.

  Next, the business hosts 5 and 6 will be described. The business host 5 includes a path information management unit 51, a path control unit 52, and a host information management table 53. Similarly, the business host 6 includes a path information management unit 61, a path control unit 62, and a host information management table 63. The path information management means 51 and 61 and the path control means 52 and 62 are realized by, for example, a CPU (not shown) executing a processing program stored in advance. The host information management tables 53 and 63 are stored in a memory (not shown).

  The path information management means 51 operates periodically and confirms the number of paths that can be used for communication with the disk array device 1 and the path that is currently used for communication. Then, the path information management unit 51 registers the number of paths that can be used for communication and the path name of the used path in the host information management table 53. The disk array device 1 also has a host information management table 18. The host information management table 53 of the business host 5 stores only information on the business host 5. Similarly, only information on the business host 6 is stored in the host information management table 63 of the business host 6. On the other hand, the host information management table 18 of the disk array device 1 stores information on all business hosts 5 and 6.

  FIG. 5 shows an example of the host information management table 18. The host information management table 18 includes a host name that is information for identifying a business host, the number of paths that can be used for I / O with the disk array device 1 on the host, and information on the currently used path (business Host side HBA WWPN value, etc.). In this embodiment, two business hosts (5, 6) are connected to four controllers (21-24) on the disk array device 1 by redundant paths via two FC switches (41, 42). It is connected. Also, among the paths connected to the business host 5, the WWNP (path name) of the HBA connected to the FC switch 41 is “A0”, and the WWNP of the HBA connected to the FC switch 42 is “A1”. Yes. Similarly, for the business host 6, the WWNP of the HBA connected to the FC switch 41 is “B0”, and the WWNP of the HBA connected to the FC switch 42 is “B1” (see FIG. 1).

  The path control means 52 (path switching means) operates periodically or when the currently used path becomes unavailable, and switches the path used by the business host 5. Specifically, the path control unit 52 switches the path to be used when the path needs to be switched, there is a path to be switched, and the path is normally connected.

  Next, the operation of the disk array device 1 according to the present embodiment will be described with reference to the flowchart shown in FIG. First, the performance management unit 11 acquires the load information of the controllers 21 to 24 and updates the performance management table 15 (step S101).

  Next, the port control unit 12 acquires path connection information (first path connection information) indicating whether or not communication is possible for the paths connected to the ports of the controllers 21 to 24, and the port state management table 16. Is updated (step S102).

  Next, the path status monitoring unit 13 acquires path connection information (second path connection information) of each path connected to all the business hosts 5 and 6, and updates the path status management table 17 (step S103). ).

 When each table of the disk array device 1 is updated in steps S101 to S103, the controller control unit 14 stops the operation of the controller based on the information of each table. First, the controller control unit 14 refers to the performance management table 15 and checks the load state of each of the controllers 21 to 24 (step S104). Then, the controller control means 14 determines whether there is an influence on the business performance when the controller is stopped (step S105).

  Specifically, when a certain controller is stopped, the controller control unit 14 determines whether or not the sum of the IOPS limit values of the remaining controllers exceeds the IOPS statistical value of the entire disk array at that time. . For example, in this embodiment, the sum of IOPS limit values when the controller 23 is stopped is 900, which exceeds 700, which is the sum of IOPS statistical values at that time. For this reason, the controller control means 14 can determine that even if the controller 23 is stopped, there is no influence on the business performance.

  When there is no influence on the business performance (step S105: Yes), the controller control means 14 refers to the port status management table 16 and the path status management table 17 to establish a connection between the disk array device 1 and the business hosts 5 and 6. The connection state is confirmed (step S106). On the other hand, when there is an influence on the business performance (step S105: No), the controller control unit 14 ends the operation.

  Next, the controller control unit 14 maintains the redundancy even if the controller identified in step S105 (the controller that does not affect the business performance) is stopped based on the port state management table 16 and the path state management table 17. It is determined whether or not to be performed (step S107). Specifically, the controller control means 14 has two or more communicable paths between the controllers 21 to 24 and the FC switches 41 and 42, and the FC switches 41 and 42 and the business host 5 (or business It is determined whether there are two or more paths that can communicate with the host 6).

  For example, in the case of this embodiment, referring to the port state management table 16, there are two paths between the other controllers 21, 22, 24 and the FC switches 41, 42 even if the port of the controller 23 is lost. There are books one by one. For this reason, the controller control unit 14 can determine that redundancy is maintained between the disk array device 1 and the FC switches 41 and 42.

  Further, referring to the path status management table 17, two paths exist between the business hosts 5 and 6 and the FC switches 41 and 42. At this time, the controller control unit 14 determines that the smaller of the number of paths between the business host and the FC switch and the number of paths between the FC switch and the controller is the path between the controller and the business host. The number can be judged.

  For example, when attention is paid to the controller 21, there are two paths between the controller 21 and the FC switches 41 and 42 and between the business host 5 and the FC switches 41 and 42. For this reason, the controller control unit 14 can determine that two or more paths exist between the controller 21 and the business host 5. The same determination can be made for the other controllers 22 to 24. In this embodiment, even if the controller 23 is stopped, there are two or more communicable paths between the controllers 21, 22, 24 and the FC switches 41, 42. There are two or more communicable paths between the FC switches 41 and 42 and the business host 5. Therefore, the controller control unit 14 can determine that the redundancy between the business host 5 and the disk array device 1 can be maintained.

  On the other hand, if there is only one path between the controller 21 and the FC switches 41 and 42, or between the business host 5 and the FC switches 41 and 42, or if there is only one path, the controller control The means 14 can determine that there is only one path between the controller 21 and the business host 5.

  If it is determined in steps S105 and S107 that the business performance requirements are satisfied and the redundancy can be maintained, the controller control unit 14 stops the controller specified in step S105 (step S108).

  Note that the update order of the performance management table 15, the port state management table 16, and the path state management table 17 (steps S101 to S103) is not limited to the above order. In addition, the order in which the controller control unit 14 determines with reference to these tables is not limited to the above order, and all conditions may be satisfied even if they are referred to and determined in different orders.

  Next, the operation of the path control means 52 of the business host 5 will be described with reference to the flowchart shown in FIG. First, the path information management means 51 confirms the number of paths that can be used for communication with the disk array device 1 among the paths connected to the business host 5, and the path currently used for communication, and manages the host information. The table 53 is updated. Similarly, the path information management means 51 confirms the number of paths that can be used for communication with the disk array device 1 among the paths connected to the business host 6 and the path that is currently used for communication. The management table 63 is updated. Further, the disk array device 1 acquires information stored in the host information management tables 53 and 63 of the business hosts 5 and 6 and updates the host information management table 18 (step S201).

  Next, the path control unit 52 operates when a path currently used by the business host 5 becomes unavailable or when the host information management table 53 is updated, and the path switching is necessary. If the path exists, switch the path to be used. Specifically, the path control unit 52 determines whether or not the currently used path is normally connected (step S202). For example, when a state where I / O to the disk array device 1 cannot be continued (failed) is detected, or when the link state of the path state management table 17 is “DOWN”, the path control unit 52 Therefore, it can be determined that the currently used path is not normally connected.

  If the currently used path is not normally connected (step S202: Yes), the path control unit 52 determines that path switching is necessary and determines whether there is another normal path. (Step S203).

  Specifically, the path control unit 52 refers to the host information management table 53 and confirms whether the number of available paths is registered as 2 or more. If there are two or more paths that can be used, the path control means 52 refers to the path status management table 17 and checks whether there are other normal paths (step S203). If another normal path exists (step S203: Yes), the path control unit 52 switches the path used by the business host 5 to one of the normal paths (step S204). When there is no path to be switched (step S202: No or step S203: No), the path control unit 52 does not perform path switching. In the above description, the path control unit 52 has been described, but the path control unit 62 of the business host 6 operates in the same manner.

  As described above, according to the configuration of the redundant system according to the present embodiment, the port controller 12 connects the path (first path) between the controllers 21 to 24 and the FC switches 41 and 42. Information (first path connection information) is acquired. Further, the path information management means 51 and 61 obtain path connection information (second path connection information) of a path (second path) between the business hosts 5 and 6 and the FC switches 41 and 42. Then, the controller control means 14 refers to the port status management table 16 and the path status management table 17, and uses the disk array device 1 and the business host 5 (or business host 6) using at least two or more communicable paths. ) To stop the operation of the controller. Thereby, even when a failure occurs in the path, it is possible to realize power saving while maintaining redundancy.

  Further, the performance management unit 11 acquires the load information of the controllers 21 to 24. And a controller control means determines the controller which can be stopped based on the sum total of the load of the present controllers 21-24, and each load of the current controllers 21-24, and stops a controller. Thereby, it is possible to save power without deteriorating business performance.

  Note that the present invention is not limited to the above-described embodiment, and can be changed as appropriate without departing from the spirit of the present invention. For example, the number of business hosts, the number of paths, the number of controllers, etc. are not limited to the configuration shown in FIG. In addition, link up and link down information is used as the first and second path connection information, but the information is not limited to this as long as the information indicates whether the path is communicable. For example, arbitrary data may be transmitted / received via each path, and the availability of communication may be checked based on the result of the transmission / reception.

  In the above embodiment, the business hosts 5 and 6 and the disk array device 1 are connected via the FC switches 41 and 42, but the present invention is not limited to this. The business hosts 5 and 6 and the controllers 21 to 24 may be directly connected. In this case, only the path connection information (first path connection information) of each path (first path) can be used to identify a communicable path and achieve redundancy. That is, as shown in FIG. 8, the disk array device includes at least a port control unit 12 that acquires path connection information of paths connected to the controllers 21 to 24, and a path state management table that stores the path connection information. 17 and controller control means 14 for stopping the controller so as to maintain redundancy based on the path connection information. With such a configuration, even when a failure occurs in a path, power saving can be realized while maintaining redundancy.

  Furthermore, arbitrary processing of the above-described redundancy system can be realized by causing a CPU (Central Processing Unit) to execute a computer program. In this case, the computer program can be stored using various types of non-transitory computer readable media and supplied to the computer. Non-transitory computer readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (for example, flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (for example, magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / W, semiconductor memory (for example, mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash memory, RAM (random access memory)) are included. The program may also be supplied to the computer by various types of transitory computer readable media. Examples of transitory computer readable media include electrical signals, optical signals, and electromagnetic waves. The temporary computer-readable medium can supply the program to the computer via a wired communication path such as an electric wire and an optical fiber, or a wireless communication path.

  In addition to the case where the function of the above-described embodiment is realized by the computer executing the program that realizes the function of the above-described embodiment, this program is not limited to the OS ( A case where the functions of the above-described embodiment are realized in cooperation with an operating system or application software is also included in the embodiment of the present invention. Furthermore, the present invention is also applicable to the case where the functions of the above-described embodiment are realized by performing all or part of the processing of the program by a function expansion board inserted into the computer or a function expansion unit connected to the computer. It is included in the embodiment.

DESCRIPTION OF SYMBOLS 1 Disk array apparatus 3 Logical disk 5, 6 Business host 11 Performance management means 12 Port control means 13 Path state monitoring means 14 Controller control means 15 Performance management table 16 Port state management table 17 Path state management table 18 Host information management table 21- 24 Controller 41, 42 FC switch 51, 61 Path information management means 52, 62 Path control means 53, 63 Host information management table

Claims (8)

A storage device having a memory and a plurality of controllers that control input / output of data to / from the memory and are connected to a plurality of first paths;
A host device connected to the storage device via a plurality of the first paths;
First connection information acquisition means for acquiring first path connection information indicating whether or not a plurality of the first paths can communicate;
Based on the first path connection information, the operation of the controller is stopped so that the connection between the storage device and the host device using the first communicable first path is maintained. Controller control means for causing
Redundant system with A plurality of relay devices connected to the plurality of controllers via the plurality of first paths and connected to the host device via a plurality of second paths;
A second connection information acquisition means for acquiring second path connection information indicating whether or not a plurality of the second paths can communicate with each other;
The controller control means connects the storage device and the relay device using at least two or more communicable first paths based on the first path connection information and the second path connection information. The operation of the controller is stopped so that the connection between the relay apparatus and the host apparatus using the second path capable of communicating at least two or more is maintained. Redundancy system. Path switching means for switching the second path used by the host device;
The path switching unit switches to the communicable second path based on the second path connection information when the second path used by the host device becomes incommunicable. Item 3. The redundant system according to item 2. Load information acquisition means for acquiring load information indicating the load of the controller;
The redundancy system according to any one of claims 1 to 3, wherein the controller control means determines the controller to be stopped based on the load information.   The redundancy system according to claim 4, wherein the load information acquisition unit determines the controller to be stopped based on load information of the controller and a preset limit value of the load on the controller. A storage device that includes a memory and a plurality of controllers that are connected to the plurality of first paths and controls input / output of data to / from the memory, and is connected to the storage devices via the plurality of first paths A redundant method of a system comprising:
Obtaining first path connection information indicating whether or not a plurality of the first paths are communicable;
Based on the first path connection information, the operation of the controller is stopped so that the connection between the storage device and the host device using the first communicable first path is maintained. Step to
A redundancy method comprising: Memory,
A plurality of controllers for controlling input / output of data to / from the memory and connected to a host device via a plurality of first paths;
First connection information acquisition means for acquiring first path connection information indicating whether or not a plurality of the first paths can communicate;
A controller that stops the operation of the controller based on the first path connection information so that the connection between the controller and the host device using the first path capable of at least two or more communication is maintained. Control means;
A storage device. A storage device that includes a memory and a plurality of controllers that are connected to the plurality of first paths and controls input / output of data to / from the memory, and is connected to the storage devices via the plurality of first paths A program for controlling a system comprising:
On the computer,
Obtaining first path connection information indicating whether or not a plurality of the first paths are communicable;
Based on the first path connection information, the operation of the controller is stopped so that the connection between the storage device and the host device using the first communicable first path is maintained. Step to
A program that executes

Images